Medical apparatus

MEDICAL SYSTEM, APPARATUS AND METHOD

RELATED APPLICATIONS 5 The present application claims the benefit of PCT Application No. filed on November 13, 2007 and of U.S. ProvisionalApplication 61/071,959 filed May 28, 2008.

BACKGROUND 10

Medical monitoring devices are routinely used in various medical settings toprovide crucial data regarding a patient’s medical condition. The monitoring devicesmay be divided into two main groups: monitoring devices that are used to monitorparameters that are a direct measure of one of the patient’s physiological functions 15 and monitoring devices that are used to monitor parameters that are an indirectmeasure of the status of a physiological function. For example, a parameter that is adirect measure for a physiological function is capnography that may be used tomeasure and provides values of the CO2 concentration in the ventilated breath, whichis a direct measure of tlie patients ventilation functioning. For example, a parameter 20 that is an indirect measurement is blood pressure, which indirectly providesinformation regarding the functioning of the heart and the cardio vascular condition ofthe patient.

Capnography is a non-invasive monitoring method used to continuously 25 measure CO2 concentration in exhaled breath. The CO2, which is a constantmetabolism product of the cells, is exhaled out of the body, and the concentration ofthe exhaled CO2, also known as end tidal CO2 (EtCO2), is an approximate estimationof the arterial levels of CO2. The measurements of the CO2 concentration in a breathcycle are performed by a capnograph, and the results are a numerical value displayed 30 also in a graphical format in the shape of a waveform named a capnogram. Thenumerical value of the results may be presented in units of pressure (mm Hg) or apercentile. The capnogram may depict CO2 concentration against total expiredvolume, but the more common capnogram illustrates CO2 concentration against time.

Analyzing the capnogram may yield valuable information about the patient’sclinical status. A normal capnogram exhibits one or more typical waveforms, eachone represents a single respiratory cycle, and deviation from the normal waveformmay hint as to the clinical situation of the patient. For example, an abnormally high 5 basal line represents re-breathing of exhaled CO2; a slow increase in CO2concentration may hint to uneven emptying of the lungs; rising in CO2 concentrationwithout reaching a plateau may hint to situations of asthma or other lower airwayobstruction; very small changes in CO2 concentration may indicate an apnea situation,and the like. In addition to displaying respiratory cycles, a trend display is also 10 available in which many individual consecutive breath cycles are compressed togetherso that changes over time may be easily distinguished, providing yet an additional aidin assessing and monitoring the patient’s ventilation and clinical profile.

Capnography is widely used today as an important tool for tracking a patient’s 15 ventilation status in various health care settings, such as an Emergency Room (ER),Operating Room (OR), Intensive Care Unit (ICU) and Emergency Medical Services(EMS). Among the clinical applications in which capnography may be used areCardiovascular (for example in CPR, shock, pulmonary embolism), Respiratory (forexample, verification of endotracheal tubing, mechanically ventilated patients, 20 conditions such as Asthma, hyperventilation, hypoventilation, apnea; Sedation (forexample during an operation) and/or Patient transport (both intra- and inter- hospital).

In addition to CO2 concentration, various other parameters may be indicative(directly or indirectly) of the ventilation (respiratory) status of a patient. Such 25 parameters may include, for example, saturation of oxygen in the blood cells andother organs, heart rate, respiration rate, breath flow rate, blood pressure, and the like.Combinations of various parameters may yield an improved indication of the clinicalcondition of the patient, in general, and of the ventilatory status of the patient, inparticular. 30 35 2

SUMMARY 5 The following embodiments and aspects thereof are described and illustrated in conjunction with systems, tools and methods which are meant to be exemplary andillustrative, not limiting in scope. In various embodiments, one or more of the above-described problems have been reduced or eliminated, while other embodiments aredirected to other advantages or improvements. 10

According to some embodiments, there is provided a method of generating apulmonary index value of a patient, which includes receiving two or more measuredpatient parameters, wherein at least one of the measured parameters originates from apulmonary sensor; and computing the pulmonary index value based on the two or 15 more measured patient parameters.

According to some embodiments, computing the pulmonary index mayinclude applying a mathematical model reflecting medical expert considerations,literature, clinical data, medical experience or any combination thereof. Computing 20 the pulmonary index may include applying a fuzzy logic, a Bayesian network, adecision tree, a neural network, a radial base function, a linear regression model, anon-linear regression model, an expert system, or any combination thereof.

According to further embodiments, the measured patient parameters maybe 25 averaged over a period of time. The period of time may be predetermined. Theperiod of time is determined based on one or more patient characteristics, on thestability of one or more measured patient parameters, or both.

According to further embodiments, the two or more measured parameters may 30 include respiration rate, a CO2 related parameter, an 02 related parameter, heart rate, an electrocardiogram (ECG), an encephalogram (EEG), blood pressure, spirometry or any combination thereof. The CO2 related parameter may include a CO2 waveform related parameter, an expired air CO2 concentration, respiratory rate or any combination thereof. The CO2 waveform related parameter may include EtCO2, 3 changes in EtC02, a slope of the increase in the CO2 concentration, a change in aslope of the increase in the CO2 concentration, time to rise to a predeterminedpercentage of a maximum value of CO2 concentration, a change in time to rise to apredetermined percentage of a maximum value of CO2 concentration, an angle of rise 5 to a predetermined percentage of a maximum value of CO2 concentration, a change inan angle of rise to a predetermined percentage of a maximum value of CO2concentration, breath to breath correlation, a change in breath to breath correlation, aCO2 duty cycle, a change in CO2 duty cycle, minute ventilation, a change in minuteventilation or any combination thereof. 10

According to additional embodiments, computing pulmonary index value mayinclude assigning a first value to a first measured patient parameter based on acomparison of the first measured patient parameter against a first reference value;assigning a second value to a second measured patient parameter based on a 15 comparison of the second measured patient parameter against a second referencevalue; and generating the pulmonary index value based on the first and the secondassigned values, wherein a weighting factor applied to the second assigned value isinfluenced by the first assigned value. 20 According to further embodiments, assigning the first value, the second value or both comprises applying a mathematical model reflecting medical expertconsiderations, literature, clinical data, medical experience or any combinationthereof. Assigning the first value, the second value or both may include applying afuzzy logic, a Bayesian network, a decision tree, a neural network, a radial base 25 function, a linear regression model, a non-linear regression model, an expert system,or any combination thereof.

According to additional embodiments, generating the pulmonary index valueincludes applying a mathematical model reflecting medical expert considerations, 30 literature, clinical data, medical experience or any combination thereof. Generatingthe pulmonary index value may include applying a fuzzy logic, a bayesian network, adecision tree, a neural network, a radial base function, a linear regression model, anon-linear regression model, an expert system, or any combination thereof. 4

According to yet additional embodiments, the method may further includereceiving a plurality of parameters.

According to yet additional embodiments, the method of generating apulmonary index value of a patient may further include applying a learning process,wherein the learning process increases reliability of the pulmonary index value. Thelearning process comprises neural network methods, a support vector machine(SVM), genetic algorithms, simulated annealing, expectation-maximization (EM),case based reasoning, or any combination thereof.

According to additional embodiments, the method of generating a pulmonaryindex value of a patient may further include receiving one or more patientcharacteristics; and computing the pulmonary index value based on the two or moremeasured patient parameters and on one or more patient characteristics. The one ormore patient characteristics may include age, weight, gender, medical condition,medication, ventilation, oxygen supply, lab tests results, blood pressure, medicalhistory, intubation or any combination thereof. The medical history may includesmoking, heart disease, lung disease, sleep apnea, a pacemaker, or any combinationthereof.

According to further embodiments, the pulmonary index value is in the rangeof 1 to 10. An increase in the pulmonary index-value may be indicative of animprovement in the condition of the patient. A decrease in the pulmonary index-valueis indicative of a deterioration of the condition of the patient.

According to yet further embodiments, computing the pulmonary index valuefurther may further include taking into account a medical significance of at least oneof the parameters.

According to additional embodiments, the method of generating a pulmonary index value of a patient may further include computing a trend of the pulmonary index-value. The method may further include computing a reliability index of the pulmonary index-value. The method may further include providing a medical recommendation. 5

According to additional embodiments, the method of generating a pulmonaryindex value of a patient may further include adjusting at least one parameter related toa patient controlled analgesia (PCA) pump, a dosage management device, a 5 ventilation device or any combination thereof, based on the pulmonary index value.

According to some embodiments, there is provided a method of computing anindex value indicative of a condition of a patient, which includes receiving two ormore measured patient parameters, wherein the measured patient parameters originate 10 from one or more sensing devices; characterizing a first measured patient parameterbased on a comparison of the first measured patient parameter against a first referencevalue; characterizing a second measured patient parameter based on a comparison ofthe second measured patient parameter against a second reference value; andcomputing the index value based on values associated with each of the characterized 15 first and second measured patient parameters, wherein a weighting factor applied tothe value associated with the second characterized measured parameter is influencedby the characterization of the first measured patient parameter.

According to additional embodiments, there is provided a method of 20 computing an index value indicative of a condition of a patient, which includesreceiving a set of measured patient parameters, wherein the set comprises at least afirst and a second measured patient parameters and wherein at least one measuredpatient parameter originates from a sensing device; assigning a first value to the firstmeasured patient parameter based on a comparison of the first measured patient 25 parameter against a first reference value; assigning a second value to the secondmeasured patient parameter based on a comparison of the second measured patientparameter against a second reference value; and generating the index value based onthe first and the second assigned values, wherein a weighting factor applied to thesecond assigned value is influenced by the first assigned value. 30

According to additional embodiments, the method may further includeassigning a third value to a third measured patient parameter based on a comparisonof the third measured patient parameter against a third reference value; and 6 generating the index value based on the first, the second and the third assigned values,wherein a weighting factor applied to the second assigned value is influenced by thefirst assigned value and the third assigned value. 5 According to yet further embodiments, there is provided a method of computing an index value indicative of a condition of a patient, which includesreceiving a set of measured patient parameters, wherein the set comprises at least twomeasured patient parameters and wherein at least one measured patient parameteroriginates from a sensing device; assigning a value to each of the at least two 10 measured patient parameters based on a comparison of the measured patientparameters against one or more reference values; and generating the index valuebased on the assigned values, wherein a weighting factor applied to at least one of theassigned values is influenced by one or more assigned values. 15 According to some embodiments, there is further provided a method of computing an index value indicative of a condition of a patient, which includesreceiving a set of measured patient parameters, wherein the set comprising at leasttwo measured patient parameters and wherein at least one measured patient parameteroriginates from a sensing device; assigning a value to each of the at least two 20 measured patient parameters based on a comparison of the measured patientparameters against one or more reference values, wherein at least one reference valueis influenced by one or more assigned values; and generating the index value based onthe assigned values. 25 According to some embodiments, computing the index value may include applying a mathematical model reflecting medical expert considerations, literature,clinical data, medical experience or any combination thereof. Computing the indexmay include applying a fuzzy logic, a Bayesian network, a decision tree, a neuralnetwork, a radial base function, a linear regression model, a non-linear regression 30 model, an expert system, or any combination thereof.

According to some embodiments, the measured patient parameters may beaveraged over a period of time. The period of time may be predetermined. The 7 period of time may be determined based on one or more patient characteristics, on thestability of one or more measured patient parameters, or both.

According to some embodiments, the two or more measured parameters may5 include respiration rate, a CO2 related parameter, an 02 related parameter, heart rate,an electrocardiogram (ECG), an encephalogram (EEG), blood pressure, spirometry, orany combination thereof. The CO2 related parameter comprises a CO2 waveformrelated parameter, an expired air CO2 concentration, respiratory rate or anycombination thereof. The CO2 waveform related parameter comprises EtC02, 10 changes in EtCO2, a slope of the increase in the CO2 concentration, a change in aslope of the increase in the CO2 concentration, time to rise to a predeterminedpercentage of a maximum value of CO2 concentration, a change in time to rise to apredetermined percentage of a maximum value of CO2 concentration, an angle of riseto a predetermined percentage of a maximum value of CO2 concentration, a change in 15 an angle of rise to a predetermined percentage of a maximum value of CO2concentration, breath to breath correlation, a change in breath to breath correlation, aCO2 duty cycle, a change in CO2 duty cycle, or any combination thereof.

According to some embodiments, tlie method of computing an index value 20 may further include applying a learning process, wherein the learning processincreases reliability of the index value. The learning process comprises neuralnetwork methods, support vector machine (SVM), genetic algorithms, simulatedannealing, expectation-maximization (EM), case based reasoning, or any combinationthereof. 25

According to some embodiments, the method of computing an index valuemay further include receiving one or more patient characteristics; and computing theindex value based on the two or more measured patient parameters and on one ormore patient characteristics. The one or more patient characteristics may include age, 30 weight, gender, medical condition, medication, ventilation, oxygen supply, lab testsresults, blood pressure, medical history, intubation or any combination thereof. Themedical history may include smoking, heart disease, lung disease, sleep apnea, apacemaker, or any combination thereof. 8

According to some embodiments, the index-value may be in the range of 1 to10. An increase in the index-value may be indicative of an improvement in thecondition of the patient. A decrease in the index-value may be indicative of adeterioration in the condition of the patient. 5

According to some embodiments, computing the index value may furtherinclude taking into account a medical significance of at least one of the parameters.

According to some embodiments, the method of computing an index value10 may further include computing a trend of the index-value. The method may furtherinclude computing a reliability index of the index-value. The method may further include providing a medical recommendation.

According to some embodiments, the method of computing an index value15 may further include adjusting at least one parameter related to a patient controlledanalgesia (PCA) pump, a dosage management device, a ventilation device or any combination thereof, based on the index value.

According to additional embodiments, there is provided an apparatus for20 generating a pulmonary index value of a patient, said apparatus includes one or moreports for receiving two or more measured patient parameters, wherein at least one ofthe measured parameters originates from a pulmonary sensor; and a processing logicadapted to compute the pulmonary index value based on the two or more measured patient parameters. 25

According to some embodiments, the processing logic may be adapted toapply a mathematical model reflecting medical expert considerations, literature,clinical data, medical experience or any combination thereof. The processing logicmay be adapted to apply a fuzzy logic, a Bayesian network, a decision tree, a neural 30 network, a radial base function, a linear regression model, a non-linear regressionmodel, an expert system, or any combination thereof.

According to further embodiments, the processing logic may further beadapted to average measured patient parameters over a period of time. The period of 9 time may be predetermined. The period of time is determined based on one or morepatient characteristics, on the stability of one or more measured patient parameters, orboth. 5 According to additional embodiments, the two or more measured parameters may include respiration rate, a CO2 related parameter, an 02 related parameter, heart rate, an electrocardiogram (ECG), an encephalogram (EEG), blood pressure,spirometry, or any combination thereof. The CO2 related parameter may include aCO2 waveform related parameter, an expired air CO2 concentration, respiratory rate 10 or any combination thereof. The CO2 waveform related parameter may includeEtCO2, changes in EtCO2, a slope of the increase in the CO2 concentration, a changein a slope of the increase in the CO2 concentration, time to rise to a predeterminedpercentage of a maximum value of CO2 concentration, a change in time to rise to apredetermined percentage of a maximum value of CO2 concentration, an angle of rise 15 to a predetermined percentage of a maximum value of CO2 concentration, a change inan angle of rise to a predetermined percentage of a maximum value of CO2concentration, breath to breath correlation, breath to breath correlation, a CO2 dutycycle, a change in CO2 duty cycle or any combination thereof. 20 According to additional embodiments, the processing logic may be adapted to compute the pulmonary index value by characterizing a first measured patientparameter based on a comparison of the first measured patient parameter against afirst reference value; characterizing a second measured patient parameter based on acomparison of the second measured patient parameter against a second reference 25 value; and generating the pulmonary index value based on values associated with eachof the characterized first and second measured patient parameters, wherein aweighting factor applied to the value associated with the second characterizedmeasured parameter is influenced by the characterization of the first measured patientparameter. 30

According to further embodiments, the processing logic may be adapted to apply a learning process, wherein the learning process increases reliability of the pulmonary index value. The learning process may include neural network methods, a support vector machine (SVM), genetic algorithms, simulated 10 annealing, expectation-maximization (EM), case based reasoning, or anycombination thereof.

According to additional embodiments, the processing logic may furtheradapted to compute the pulmonary index value based on the two or more measuredpatient parameters and on one or more patient characteristics. The one or morepatient characteristics may include age, weight, gender, medical condition,medication, ventilation, oxygen supply, lab tests results, blood pressure, medicalhistory, intubation or any combination thereof. The medical history may includesmoking, heart disease, lung disease, sleep apnea, a pacemaker, or any combinationthereof.

According to further embodiments, the pulmonary index-value may be in therange of 1 to 10. An increase in the pulmonary index-value may be indicative of animprovement in the condition of the patient. A decrease in the pulmonary index-valuemay be indicative of a deterioration in the condition of the patient.

According to further embodiments, the processing logic may be furtheradapted to compute the pulmonary index value according to a medical significance ofat least one of the parameters. The processing logic may be further adapted tocompute a trend of the pulmonary index-value. The processing logic is furtheradapted to compute a reliability index of the pulmonary index-value. The processinglogic is further adapted to provide a medical recommendation.

According to yet additional embodiments, the apparatus may further include agraphic display of the pulmonary index value. The apparatus may further include acontroller adapted to adjust at least one parameter related to a patient controlledanalgesia (PCA) pump, a dosage management, a ventilation device or anycombination thereof, based on the pulmonary index value.

According to some embodiments, there is provided an apparatus for generatingan index value indicative of a condition of a patient, said apparatus include a portadapted to receive two or more measured patient parameters, wherein the measured 11 08/000894 patient parameters originate from one or more sensing devices; and processing logicadapted to: (a) characterize a first measured patient parameter based on a comparisonof the first measured patient parameter against a first reference value; (b) characterizea second measured patient parameter based on a comparison of the second measured 5 patient parameter against a second reference value; and (c) computing the index valuebased on values associated with each of the characterized first and second measuredparameters, wherein a weighting factor applied to the value associated.with the secondcharacterized measured parameter is influenced by the characterization of the firstmeasured patient parameter. 10

According to additional embodiments, the processing logic may be adapted toapply a mathematical model reflecting medical expert considerations, literature,clinical data, medical experience or any combination thereof, The processing logicmay be adapted to apply a fuzzy logic, a Bayesian network, a decision tree, a neural 15 network, a radial base function, a linear regression model, a non-linear regressionmodel, an expert system, or any combination thereof.

According to further embodiments, the processing logic may be adapted toaverage measured patient parameters over a period of time. The period of time is 20 predetermined. The period of time is determined based on one or more patientcharacteristics, on the stability of one or more measured patient parameters, or both.

According to some embodiments, the two or more measured parameters mayinclude respiration rate, a CO2 related parameter, an 02 related parameter, heart rate, 25 an electrocardiogram (ECG), an encephalogram (EEG), blood pressure, spirometry, orany combination thereof. The CO2 related parameter may include a CO2 waveformrelated parameter, an expired air CO2 concentration, respiratory rate or anycombination thereof. The CO2 waveform related parameter may include EtCO2,changes in EtCO2, a slope of the increase in the CO2 concentration, a change in a 30 slope of the increase in the CO2 concentration, time to rise to a predeterminedpercentage of a maximum value of CO2 concentration, a change in time to rise to apredetermined percentage of a maximum value of CO2 concentration, an angle of riseto a predetennined percentage of a maximum value of CO2 concentration, a change inan angle of rise to a predetermined percentage of a maximum value of CO2 12 concentration, breath to breath correlation, breath to breath correlation, a CO2 duty cycle, a change in CO2 duty cycle or any combination thereof.

According to further embodiments, the processing logic may be adapted to5 apply a learning process, wherein the learning process increases reliability of thepulmonary index value. The learning process may include neural network methods, asupport vector machine (SVM), genetic algorithms, simulated annealing, expectation- maximization (EM), case based reasoning, or any combination thereof. 10 According to additional embodiments, the processing logic may further be adapted to compute the pulmonary index value based on the two or more measuredpatient parameters and on one or more patient characteristics. The one or morepatient characteristics may include age, weight, gender, medical condition,medication, ventilation, oxygen supply, lab tests results, blood pressure, medical 15 history, intubation or any combination thereof. The medical history may includesmoking, heart disease, lung disease, sleep apnea, a pacemaker, or any combinationthereof.

According to further embodiments, the index-value may be in the range of 1 to 20 10. An increase in the index-value may be indicative of an improvement in the condition of the patient. A decrease in the index-value may be indicative of adeterioration in the condition of the patient.

According to further embodiments, the processing logic may be further 25 adapted to compute the index value according to a medical significance of at least oneof the parameters. The processing logic may be further adapted to compute a trend ofthe index-value. The processing logic is further adapted to compute a reliability indexof the index-value. The processing logic is further adapted to provide a medicalrecommendation. 30

According to yet additional embodiments, the apparatus may further include a graphic display of the index value. The apparatus may further include a controller adapted to adjust at least one parameter related to a patient controlled analgesia (PCA) 13 pump, a dosage management, a ventilation device or any combination thereof, based on the pulmonary index value.

In addition to the exemplary aspects and embodiments described above,5 further aspects and embodiments will become apparent by reference to the figures and by study of the following detailed descriptions. 10 BRIEF DESCRIPTION OF THE FIGURES:

Fig. 1A - A graph depicting the CO2 medical significance level;

Fig. IB- A graph depicting the respiration rate medical significance level; 15 Fig. 1C— A graph depicting the SpO2 medical significance level;

Fig. ID - A graph depicting the heart rate medical significance level;

Fig . 2 - An exemplary graph depicting membership as a function of age, when using a fuzzy logic interface model, according to some embodiments;

Fig. 3 - An exemplary graph depicting the PI value as a function of the level of 20 risk;

Fig. 4A - An exemplary illustration of a method of calculating an index value,according to some embodiments;

Fig. 4B - An exemplary illustration of a Bayesian network for calculating anindex value, according to some embodiments; 25 Fig. 4C - An exemplary illustration of a decision tree graph structure for calculating an index value, according to some embodiments;

Fig. 4D - An exemplary illustration of a feed forwards neural network forcalculating an index value, according to some embodiments;

Fig. 5 - An exemplary block diagram scheme of a learning model, according 30 to some embodiments;

Fig. 6A - A block diagram illustration of graphical user interfaces, accordingto some embodiments;

Fig. 6B-D - Exemplary illustrations of a graphical user interface, according tosome embodiments; 14

Fig. 7 - A block diagram of a medical monitoring system, according to someembodiments; and

Fig. 8 - A schematic illustration of a system, according to some embodiments. 5

DETAILED DESCRIPTION

In the following description, various aspects of the invention will be described.For the purpose of explanation, specific configurations and details are set forth in 10 order to provide a thorough understanding of the invention. However, it will also beapparent to one skilled in the art that the invention may be practiced without specificdetails being presented herein. Furthermore, well-known features may be omitted orsimplified in order not to obscure the invention. 15 As referred to herein, the terms "user", "medical user", "health care provider" and "health care professional" may interchangeably be used. The terms may includeany health care provider who may treat and/or attend to a patient. A user mayinclude, for example, a nurse, respiratory therapist, physician, anesthesiologist, andthe like, however, a user may also include a patient, 20

As referred to herein, the terms "device", "monitoring device" and "medicaldevice" may interchangeably be used.

As referred to herein, the terms “patient” and subject” may interchangeably be 25 used and may relate to a subject being monitored by any monitoring device for anyphysical-condition related parameter and/or health related parameter.

As referred to herein, the terms ordinary, normal, typical, standard andcommon may interchangeably be used. 30

As referred to herein, the terms “condition-index-value”, “health index” and “index value” may interchangeably be used. 15

As referred to herein, the term “pulmonary” includes relating to, affecting, oroccurring in the lungs.

As referred to herein, the term “respiratory” relates to the system that consists5 of or includes the airways, the lungs, and the respiratory muscles that mediate the movement of air into and out of the body.

As referred herein, the term “ventilatory” relates to ventilation (physiology),the movement of air into and outside the body. 10

According to some embodiments, the term ’’measured patient parameter/s”(such as a first, a second and/or any other measured patient parameter) may refer toany measurable or sensed parameter relating to a patient, such as, for example, butlimited to: breath related parmeters, such as, for example, respiratory rate, CO2 15 related parameters, and the like; O2 related parameters, such as, for example, SpO2,O2 saturation, and the like; heart related parameters, such as, for example, heart rate,ECG, blood pressure, and the like; neurological parameters, such as, for example,EEG; spirometry related parameters, such as, for example, FEV1, FVC, and the like;and the like. 20

According to some embodiments, the term “reference value” (such as a first, asecond and/or any other reference value) may include a value, a range of values ormay be defining a portion of a range of values. 25 According to some embodiments, the term ’’pulmonary index value” may refer to a pulmonary index and/or a respiratory index. The term ’’pulmonary index value”may further relate to a respiratory and cardiac index and/or to a pulmonary andcardiac index. 30 According to some embodiments, the term ’’pulmonary sensor” may include any device, sensor, system, or the like, adapted to obtain, determine, sense and/ormeasure a pulmonary and/or respiratory related parameter, such as a capnograph,spirometer, flow meter, oximeter, an acoustic measurment device, or any combinationthereof. 16

According to some embdiments, the terms “calculated” and “computed” mayinterchangeably be used. 5 Currently, in most health care settings, patient related parameters (data) are collected on line (or by other means) and may provide various health care providers,such as a nurse, a physician, a respiratory therapist, an anesthesiologist and the like,with information regarding the patient’s status. The information routinely presentedmay include various types of information regarding various parameters that may be 10 sensed by various sensors. Viewing and interpreting the information presented maysometimes be a cumbersome, complicated and time consuming task for the health careprovider.

According to some embodiments, there is a need to analyze the various patient 15 related parameters (data) that are collected on line, in real time, and provide the healthcare providers with a more comprehensive, comprehensible, meaningful, intuitive,clearer and useful information about the patient’s status. The data collected may bedetermined/calculated/computed and the information to the health care provider maybe provided in the form of a condition-index-value that is directly related to the 20 clinical condition of the patient. The condition-index-value may bedetermined/calculated/computed based upon various parameters of a patient that maybe sensed/measured by appropriate sensors. The various parameters, according towhich the condition index value is determined, may each have different units andoccasionally, different units may be used for the same parameter. Moreover, the 25 absolute values of the parameters may not always be intuitive for understanding/interpretation and neither are they linearly proportional to severity ofthe condition. In addition, some parameters may have different meanings as to thecondition of the patient when increasing and/or when decreasing, that is, for someparameters, decrease in the value indicates improvement while with other parameters, 30 decrease in value may indicate deterioration of the patient condition. This furtherdemonstrate the importance of a condition index value, which integrates variousparameters that may be measured in different units and may have different meaningsinto one comprehensible index value, which may be indicative of the absolute patientcondition. Providing the condition-index value to the health care provider may result 17 in a clear indication for the health care provider to realize when more medicalattention is required for a given patient and for a given patient’s condition. Since thecondition-index value and the indications resulting therefrom may be deduced fromseveral parameters, the sensitivity of monitoring the patient’s condition may be 5 increased, and earlier warnings with earlier intervention may be anticipated.

According to further embodiments, the condition-index-value may bedetermined/calculated/computed/determined by a device, such as, for example, amonitoring device. The monitoring device may include one or more sensors that may 10 be used to sense and/or measure, and/or calculate and/or determine various healthrelated parameters. The monitoring device may include any known medicalmonitoring device, such as, for example, capnograph, oxymeter, spirometer, heart ratesensors, blood pressure sensors, ECG, EEG, Ultrasound, and the like, and/or anycombination thereof. Based on the various measurements, the device may calculate 15 the condition-index-value, and display the index alone or in combination with thevarious patient parameters that may be sensed/measured by appropriate sensors. Inaddition, the device may also provide medical recommendations to the user based onthe analysis of the collected patient data. Moreover, the device may track and notifythe health care provider regarding changes over time of the patient’s condition. For 20 example, the device may notify the health care provider if the patient’s condition isstable, improving or deteriorating.

According to some embodiments, the condition-index-value may be a unit-lessvalue in any predetermined range, such as, for example, in the range of 1 to 100. For 25 example, condition-index-value may in the range of 1 to 10, wherein 10 indicates thebest condition, and 1 indicates the worst condition. Within the range of 1 to 10, subranges (subdivisions) may be assigned. For example, a sub-range from 8 to 10 maybe indicative of a stable, normal condition, where no intervention is needed. A sub-range of 6-7 may be indicative to the health care provider that more attention is 30 needed patient re-evaluation is recommended. A sub range of below 5 may indicateto the health care provider that intervention is needed and patient re-evaluation maybe necessary. In addition, the various sub-ranges of the condition-index-value may beassigned different graphical signs, when displayed to the health care provider. Thedifferent graphical signs may include, for example, different colors, different units, 18 different letters, and the like. For example, for condition-index-value in the sub-rangeof 8 to 10, the value may be colored green, for condition-index-value in the sub-rangeof 5 to 7, the value may be colored yellow, and for condition-index-value in the sub-range of below 5, the value may be colored red. In addition, various other visual 5 indicators may also be used to indicate changes that may be correlated with knownphysical conditions, such as, for example, up and down arrows that may indicate, forexample, an increase or decrease, respectively, in one or more measured parameters.

According to some embodiments, the condition index value may be calculated 10 by various means, such as, for example, by use of mathematical equations, algorithms,formulas, and the like, that may take into consideration one or more of the values orderivatives of the values of the parameters that are being measured by the monitoringdevice. 15 According to further embodiments, the change of the condition-index-value over time (referred to herein as condition-index-value trend) may be displayedgraphically. The graphic display may exhibit the condition-index-value trended overthe last V (time units) of monitoring. For example, n may be any time period in therange of 5 minutes to 12 hours. This display may be used to indicate the patient’s 20 status, such as, for example: stable, improving, deteriorating, as well as providing adepiction of the rate and change of the patient’s status. Displaying of the condition-index-value trend may simplify the assessment of the changes in the condition of thepatient as compared to assessing the patient condition based on the trend of theindividual parameters. When looking at the trends of the individual parameters, it 25 may not be easy and intuitive to determine the patient’s status and change in status,without taking into consideration the absolute values of the individual parameters andtheir interactions.

According to further embodiments, the condition-index-value trend may be 30 depicted as a graphic display of the condition-index-value over time. The durationperiod of the trend may be chosen to be over any time period in the range of, forexample between 5 minutes to 12 hours of the last measurements. The resolution ofthe graphical display may change accordingly in correlation to the selected timeperiod. 19

According to further embodiments, an index of reliability (referred to hereinalso as "reliability index" or "RI") may also be determined. The index of reliabilitymay provide a measure of the reliability of the data and more specifically, the 5 reliability of the condition-index-value. For example, the reliability index may beused to predict and anticipate artifacts. The reliability index may be determined, forexample, by analysis of the. CO2 waveforms, as depicted by a capnogram. If breathflow is also measured, its waveforms may also be used for this purpose. The use ofbreath flow measurements may refine and improve the index of reliability. Breath 10 flow waveforms strongly complement the waveforms created by the CO2measurement, since both measurements represent essentially the same event, which isthe breath cycle. While the breath flow relates to the envelope of the waveform, theCO2 relates to the CO2 concentration within the envelope. Using both parametersmay better reveal and uncover what is a measurement noise, artifact, and the like. 15

According to additional embodiments, a pause frequency parameter may bedetermined. This parameter may include a measure of events wherein no breathing isdetected, over a period of time. The events of lack of breathing may include, forexample, pause and apnea events, and the pause frequency parameter may include a 20 measure of a patient's pause and apnea events over a period of time. The pausefrequency parameter may be determined/calculated from the CO2 waveform, asobtained by the capnogram. A pause event may be defined, for example, as anyinhalation stage that persists for longer than any number of seconds in the range of,for example, 5 to 40 seconds (such as for example 20 seconds), and proceeds after an 25 exhalation period lasting less than any number of seconds in the range of, forexample, 5 to 20 seconds (such as, for example, 10 seconds). The time periods maybe determined, for example, according to the average time of the last three exhalationcycles. Such determination of a pause event may be used to exclude a slow, rhythmicbreathing pattern from being defined as a group of pause events. In addition, a 30 maximum time out of, for example, 100 seconds may be determined. If a pause isdetected, a new pause can only be counted if at least three new, valid breath cycleswere detected beforehand. Thus, the Pause Frequency Parameter may be defined bythe number of pause events per period of time (such as, for example, an hour). Thepause frequency may be updated at any time intervals, such as, for example, every 5 20 minutes, after the period of 1 hour. The values of the pause frequency may further bestored and used for the display of the pause frequency trend, wherein the trend datarepresent the change of the pause frequency over time. According to some exemplaryembodiments, during the first hour (when insufficient data has accumulated), a value 5 may be provided and updated, for example, every 15 minutes until 1 hour has beenreached (wherein during this time period the frequency is determined/calculated as ifit was determined/calculated for 1 hour). During this time period an indicationshowing that the pause frequency is still based on a shorter period than 1 hour may bedisplayed. Since the health care provider, such as a nurse, may not be constantly 10 present next to the patient and/or the monitoring device, and may not constantly track (monitor) the patient condition, a parameter, such as the pause frequency parameter,which is a periodic type effect, may not easily be observed by the health care providerif not otherwise tracked by the monitoring device. In addition, according to furtherembodiments, the pause amplitude parameter may also be determined. The pause 15 amplitude parameter may be determined by measurements of the time length (such as,for example, in the range of 5 to 60 seconds) of each of the detected pause events andthe dispersion of the time lengths of the pause events over a period of time (such as,for example, over a time period of 60 minutes). 20 According to some embodiments, the device may further include and display medical recommendations to the health care professionals. The medicalrecommendations may be deduced from analysis of at least some of the individualparameter values and patterns and comparison of the measured values and patterns tothe known ranges and patterns of the individual parameters. These recommendations 25 may be displayed in addition to the indications derived from thedetermined/calculated condition-index-value.

According to some embodiments the medical device may include a userinterface that may allow the user to select the data to be displayed and to control 30 various operating parameters. Moreover, different displays may be included toaccommodate different needs of the different users (such as a nurse, a physician, ananesthesiologist, and the like). Allowing the user to change the view of the data maypermit the user to toggle through the different levels of information for furtherevaluation of a condition. For example, the basic screen may display the condition- 21 index-value and the condition-index-value trend data. Changing to the next displaymay reveal the actual (measured) data values and the trends of the values that relate tothe parameters from which the condition-index-value is determined/calculated.Further toggling the display may provide the pause frequency and other related 5 analysis and calculation. The use of the various displays may also allow the user tofocus on the parameters that caused an indication of an event and/or recommendationto the user.

According to further embodiments, the user interface may also allow the user 10 to enter information that is characteristic for each patient. The use of characteristicpatient information is necessary to allow accuracy of the various measurements andcalculations. Such information may include, for example age, weight, height, sex, andthe like, of the specific patient. In addition, classification detection means of variouspatients may be utilized, wherein the classification may be based on parameters such 15 as, for example, age group, weight group, sex, and the like. Using such classificationmay allow the monitoring device to correct its settings to be appropriate for thatrelevant patient type and environment.

According to some exemplary embodiments, the condition-index-value may 20 represent the respiratory status and/or the pulmonary status and/or the cardiac statusof the patient. This may be accomplished by providing an index value that isdetermined/calculated based upon one or more of various respiratory, pulmonaryand/or cardiac parameters, such as, for example, but not limited to: CO2 relatedparameters; O2 related parameters; EtCO2; CO2 waveform related parameters, such as, 25 for example, changes in ETCO2, CO2 duty cycle, inhalation to exhalation ratio, aslope of the increase in the CO2 concentration, a change in a slope of the increase inthe CO2 concentration, time to rise to a predetermined percentage of a maximumvalue of CO2 concentration, an angle of rise to a predetermined percentage of amaximum value of CO2 concentration, breath to breath correlation, CAP-FEV1 30 (forced expiratory volume over 1 sec obtained from at least one capnographicmeasurement, a measure of flow), CAP-FEV1/FVC, and the like; an expired air CO2concentration; breathing related parameters; heart function related parameters;neurological parameters; systemic perfusion related parameters; visual parameters;FEV1 (forced expiratory volume over 1 sec, a measure of flow); FVC (forced vital 22 capacity, a measure of volume); blood pressure; NI blood pressure; systolic todiastolic ratio; respiratory rate; breath flow rate; spirometry readings; O2 saturation;SpO2; blood pressure; blood gases; heart rate; Electrocardiogram (ECG);Electroencephalogram (EEG); Ultrasound measurements, such as heart echogram, and 5 the like, or any combination thereof.

As referred to herein, the term EtCO2 relates to End tidal CO2. The CO2 isexhaled out of the body and the concentration of the exhaled CO2, also known as endtidal CO2 (EtCO2) is an approximate estimation of the alveolar CO2 pressure and thus 10 of the arterial levels of CO2. The measurements of the CO2 concentration in a breath cycle are performed by a capnograph, and the results are a numerical value displayedalso in a graphical format in the shape of a waveform named a capnogram. Thevalues of EtCO2 may be measured in units of volume or pressure, such as, forexample, mmHg. 15

As referred to herein, the term SpO2 relates to the saturation of peripheraloxygen. It is a measurement of the amount of oxygen attached to the hemoglobin inred blood cells in the circulatory system. SpO2 values are generally given as apercentage (for example, normal value is above 96%). SpO2 may be monitored and 20 measured by various monitors, such as, for example, a Pulse Oximeter.

As referred to herein, the term Respiration Rate (RR) is defined as the numberof breaths taken in a minute, and it may change under various physiological andmedical conditions. The rate may be abnormally high (tachypnea), abnormally low 25 (bradypnea) or non-existent (apnea).

As referred to herein, the term Heart (Pulse) Rate (HR) relates to the numberof heart pulses (beats) in a minute. Pulse rate is usually considered to be acombination of left ventricular stroke volume, ejection velocity, the relative 30 compliance and capacity of the arterial system, and the pressure waves that result.from the antegrade flow of blood and reflections of the arterial pressure pulsereturning from the peripheral circulation, and some or all of which may be effected byCO2. For example, in acute pulmonary edema secondary to left ventriculardecompensation, there is frank respiratory failure. Also, the effect on the heart 23 following drug administration, such as, for example, morphine, will lead to a fall inpC>2 (partial O2 pressure in blood) and a rise in pCO2 (partial pressure CO2 in blood).Thus, heart rate may be used to indicate severity of respiratory/pulmonary status. 5 As referred to herein, the term "hypoventilation" relates to a state of respiratory depression that may occur when ventilation is inadequate to performneeded gas exchange. Hypoventilation may cause an increased concentration ofcarbon dioxide and respiratory acidosis. Hypoventilation may be caused by variousmedical conditions and/or by use of some drugs and medicines. 10

As referred to herein, the term "hyperventilation" relates to a state of breathingfaster and/or deeper than necessary, thereby reducing the carbon dioxideconcentration of the blood below normal. 15 According to some exemplary embodiments, the condition-index-value may be a Pulmonary/Respiratory Index value (also referred to herein as "PI"). The PIindex may represent a measurement of the patient's respiratory and/or pulmonaryand/or cardiac status. It may be deduced from various measured parameters, such as,for example: EtCO2, Respiration Rate, SpCti and Heart Rate. The individual 20 parameters, according to which the PI is calculated, may each have different units,and occasionally, different units may be used for the same parameter, such as, forexample, for EtCCh that can be defined in units of, for example, kPas, mmHg or Vol(percent). Moreover, the absolute values of the parameters may not always beintuitive for understand ing/interpretation and neither are they linearly proportional to 25 severity of the condition. In addition, some parameters may have different meaningsas to the condition of the patient when increasing and/or when decreasing, that is, forsome parameters, decrease in the value indicates improvement while with otherparameters, decrease in value may indicate deterioration of the patient condition.These and other reasons, demonstrate the importance of a condition index value, such 30 as the PI, which integrates various parameters that may be measured in different unitsand may have different meanings into one comprehensible index value, which may beindicative of the absolute patient condition in general and on the respiratory and/orpulmonary and/or cardiac condition of the patient in particular. The PI may be a unit-less value in the range of 1 to 10, wherein 10 indicates the best condition, and 1 24 indicate the worst condition. Within the range of 1 to 10, sub ranges (subdivisions)may be assigned. For example, a sub-range from 8 to 10 may be indicative of astable, normal condition, where no intervention is needed. A sub-range of 6-7 may beindicative for the health care provider that more attention is needed. A sub range of 5 below 5 may indicate to the health care provider that intervention and/or patient re-evaluation and/or a change in therapy is recommended. In addition, the various sub-ranges of the condition-index-value may be assigned different graphical signs, whendisplayed to the health care provider. The different graphical signs may include, forexample, different colors, different units, different letters, and the like. For example, 10 for condition-index-value in the sub-range of 8 to 10, the value may be colored green, for condition-index-value in the sub-range of 5 to 7, the value may be colored yellowand for condition-index-value in the sub-range of below 5, the value may be coloredred. In addition, various other visual indicators may also be used to indicate changesthat may be correlated with known medical conditions, such as, for example, up and 15 down arrows that may indicate, for example a state of hyperventilation andhypoventilation, respectively,

According to some embodiments, the PI index value may be determined byvarious ways, using various calculation methods and various algorithms, as further 20 detailed below herein. Generally, the PI may be deduced from various parameters andmay be assigned the highest value, for example “10”, when the individual values ofthe various parameters are well within their respective normal ranges. The PI valuemay decrease below “10” when the value of one or more of the individual parameterschanges from the normal respective ranges for those parameters. The decrease of the 25 PI value may be sharper when several individual parameters change together.

According to some embodiments, the PI value may be updated continuously,and it may be determined/calculated from an average of the values of the parametersthat are used to produce the PI value. In addition, the averaging time used for the 30 determination of the PI value may also be adaptive. For example, if there is an erraticmeasurement, the average time may increase. The erratic characteristics used fordeciding the averaging time may result, for example, from the respiratory rate values,the CO2 waveform, and any other suitable parameter that is used for the calculation ofthe PI. 25

According to further embodiments, the PI value may also be indicative ofconditions such as hypoventilation and hyperventilation. When the PI is indicative ofthese conditions, an appropriate additional indicative signaling may be displayed, 5 such as, for example, an upward arrow (indicative of hyperventilation) and downwardarrow (indicative of hypoventilation). The decision as to whether the patient’s statusis in either hypoventilation or hyperventilation may be based, for example, upon theRespiratory Rate, when respiratory rate is one of the measured parameters.

10 According to some embodiments, there are various methods to calculate the PI value. Generally, various parameters may be measured and used for the calculation ofthe PI value. According to some exemplary embodiments, the PI may bedetermined/calculated based on measuring the values of at least one of the parameters:EtCO2, Respiration Rate, SpO2 and Heart Rate. The real time values of these 15 parameters may be measured continuously. For example, an adaptive running averagemay be collected for all of the 4 measured parameters (average EtCO2, averagerespiratory rate, average heart rate and average SpO2). This adaptive running averagemay be determined/calculated by collecting the data displayed by the monitors thatmeasure the parameters each second, and averaging over a period of time (as 20 explained below). In this way, the determined/calculated average takes intoconsideration not only the values collected over the last “x” number of seconds, butalso the length of time the value was displayed. The PI may then bedetermined/calculated using these average values. The averaging time period may bedetermined/calculated and defined using an adaptive type algorithm. For example, the 25 default time period may be in the range of 5 to 60 seconds, such as, for example, 30seconds. The time may increase in steps of, for example, 2 to 30 seconds, such as, forexample, in steps of 15 seconds; and the maximum period of time may be in the rangeof 5 to 180 seconds, such as, for example, 90 seconds, 120 seconds, and the like. Inorder to evaluate if the data is stable or erratic, information that may be used to 30 determine if averaging time is to be increased or decreased, respectively, theparameter value of respiratory rate may be used. The standard deviation of therespiratory rate over the last predetermined period of time (such as, for example, 30seconds) may be continuously measured. If the standard deviation of the respiratoryrate value is below a predefined threshold, then the averaging period does not change. 26

If the standard deviation of the respiratory rate over the last predetermined period oftime is above the predetermined threshold, then the averaging period may beincreased (for example, by 15 seconds). 5 According to some embodiments, the PI may be determined/calculated by using mathematic calculations. The calculation may be based on the measuring thevalues of at least one of the parameters: EtCO2, Respiration Rate (RR), SpO2 andHeart Rate (HR), and the determined/calculated average of those parameters, asdetailed hereinabove. The calculations may relay on known defined ranges values for 10 each of the measured parameters (in correlation with the patient characteristics, asdetailed below). Meeting predefined conditions of the various measured parametersvalues result in an appropriate determined/calculated PI value. For example, and asfurther detailed in Example 1, when the following conditions are realized, the PI may10 or 9 and no arrows, indicative of changes in condition are displayed: If RR is >12 15 &amp; < 28 and EtCO2 is > 28 &amp; < 44 and SpO2 is > 94%, Then: PI = 10. IfRRis>12&amp; < 28 and EtCO2 is > 28 &amp; < 44 and SpO2 > 90% &amp; < 94%, Then: PI == 9.

According to some embodiments, the PI may be determined/calculated byusing mathematic calculations. The calculations may be based on multiplication of 20 the medical significance level (risk/probability level) that is associated with each ofthe measured parameters. The medical significance level of each of the measuredparameters may be determined by creating a graph, indexing tables, and the like,which correlate the value of the parameter with an ordinary (standard/typical,common/normal) level of a physical condition, such as, for example ventilatory 25 condition (such as breathing, respiration, exhaling, inhaling). The ordinary level ofthe physical condition may be in the range of 0 to 1, wherein 1 signifies the bestphysical condition and 0 signifies the worse physical condition. Thus, for example,mathematical functions may be provided for each of the measured parameters, wherethe maximum value of 1 may relate to a physical condition that is normal, and a 30 minimum value of 0 may relate to a physical condition that is worst, such as forexample, when no ventilatory performance is detected at all. Fig. 1A illustrates agraph which depicts the medical significance (risk levels) of the EtCO2 parameter. Asshown in Fig. 1 A, the Y-axis is the ordinary level, on the scale of 0 to 1. The X-axis isthe level of EtCO2 in units of mmHg. The medical significance curve depicts the 27 correlation curve between the ordinary level and the level of EtCO2 and tliedetermined/calculated curve that best correlates to the medical significance curve. Ascan be deduced from the graph, at low EtC02, the ordinary level value decreases tozero, indicative of moving towards apnea; at high EtC02, although indicative of a 5 dangerous condition (that may cause the blood to change its alkali level which mayconsequently many essential chemical processes), the ordinary level value does notfall to zero. The equation of the exemplary medical significance curve, which isillustrated in Fig. 1A may be described by the equation: Yj = 8E-06x3 - 0.0015x2 +0.0724x - 0.0496. Fig. IB illustrates an exemplary graph which depicts the medical 10 significance levels of the respiration rate parameter. As shown in Fig. IB, the Y-axisis the ordinary level, on a scale of 0 to 1. The X-axis is the respiratory rate in units ofnumber of breaths in a minute. The medical significance curve depicts the correlationcurve between the ordinary level and respiratory rate and the determined/calculatedcurve that best correlates to the medical significance curve. The equation of the 15 exemplary medical significance curve, which is illustrated in Fig. IB, may bedescribed by the equation: Y2 = 4E-05x3 - 0.0043x2 + 0.123 lx - 0.0378. Fig. 1Cillustrates an exemplary graph which depicts the medical significance levels of theSpO2 parameter. As shown in Fig. 1C, the Y-axis is the ordinary level, on a scale of 0to 1. The X-axis is the SpO2 percentile. The medical significance curve depicts the 20 correlation curve between the ordinary level and the SpO2 and thedetermined/calculated curve that best correlates to the medical significance curve.The exemplary equation of the medical significance curve which is illustrated in Fig.1C may be described by the equation: Y3 = -2E-05x3 + 0.004x2 - 0.2778x + 6.1214.Fig. ID illustrates an exemplary graph which depicts the medical significance levels 25 of the heart rate parameter. As shown in Fig. ID, the Y-axis is the ordinary level, on a scale of 0 to 1. The X-axis is the heart rate that is measured in number of beats in aminute. The medical significance curve depicts the correlation curve between theordinary level and the heart rate and the determined/calculated curve that bestcorrelates to the medical significance curve. The equation of the exemplary medical 30 significance curve which is illustrated in Fig. ID may be described by the equation:Y4— 8E-07x3 - 0.0003x2 + 0.0406x - 0.4389. Similarly to creating and calculatingthe correlation equations, indexing tables may also be used to correlate between theordinary level and the value of each of the measured parameters. 28

According to some embodiments, the PI may be determined/calculated bymultiplying the medical significance factors of each of the measured parametersobtained by tlie equations detailed above herein, and multiplying the result by 10, toget an IP value in the range of 1 tolO. The equation may be described as: PI=Yj* Y2* 5 Y3* Y4. According to some exemplary embodiments, the HR medical significance value (Y4) is introduced to the calculations only if the medical significance value ofone of the other parameters is less than 0.8. In addition, the equations presentedhereinabove are valid up to the following maximal values of the individualparameters: the EtCO2 value reaches a value of about 90mmHg, the respiratory rate 10 reaches a value of about 50bpm and SpO2 reaches 50%. Above or below these valuesa default value of 0.2 may be used.

According to additional embodiments, if the determined/calculated index isbelow 8, an upward arrow indication may be included if the respiration rate is greater 15 than a predetermined number of beats per minute (BPM), such as, for example, 24BPM. A downward arrow indication may be included if the respiration rate is belowa predetermined number of BPM, such as, for example, 12 BPM. If the respirationrate is within a predetermined range, such as, for example between 12 and 24 BPM,no arrow indication is provided. 20

According to further embodiments, the PI may be determined/calculated bythe use of various algorithms that may be used to create a mathematical model for theindex that may reflect the “real time” assessment of a health care provider on thepatient’s pulmonary/respiratory status. In this respect, “real time” may be defined as 25 being “in-time” for the health care provider to respond to a medical situation. “Realtime” may be, for example, in the range of seconds (for example, 0 to 120 seconds), inthe range of minutes (for example 1 to 10 minutes), and the like. Some of therequirements from such an algorithm or method are: to express correctly theconsensus assessment of various health care providers, such as medical experts, 30 physicians, nurses, and the like, wherein the consensus assessment may be determinedby a weighted and/or un-weighted average. By weighted average, for example, aphysician’s assessment may be weighted more than that of a respiratory therapist. Forexample, by weighted average, each health care provider may “weight” their opinion,for example by adding a confidence value to their assessment.; easy and flexible 29 implementation of the algorithm (that is, simplify the fine-tuning of the mathematicalmodels); and fast calculations. By implementing the algorithm, the PI may thus beused in effect to replace the health care provider and his decision malting approach. 5 According to some embodiments, the algorithms used to create or be used within the mathematical model for the index may include, for example, a fuzzy logicinference that may be built, composed or enhanced using the health care providersknowledge and interpretation on the anticipated index value. For example, the indexPI values may be given for any number and/or any combination of various types of 10 medical parameters. According to some exemplary embodiments, the parametersused may include, for example, parameters, such as: EtCO2, Respiratory Rate (RR),Heart Rate (HR), SpO2, blood pressure, spirometry, relative flow parameters, CO2waveform patterns, blood gas, and the like.

15 According to additional exemplary embodiments, in order to determine the PI value based on, for example, four parameters (EtCO2, RR, HR and SpO2), aquestionnaire and/or real patient log files may be sent to various health care providers(such as, nurses, respiratory therapists, physicians and anesthesiologists). Thequestionnaire may include a set of various cases with different parameter values. The 20 health care providers may then be asked to assign a PI value according to a predefinedcode. For example, the predefined code may implicate that an index value of 10implies a perfectly healthy normal condition. For example, the predefined code mayimplicate that an index value of 8-9 implies a normal condition. For example, thepredefined code may implicate that an index value of 7 implies an understandable 25 condition, which requires, however, more attention. For example, the predefined codemay implicate that an index value of 5-6 implies that it is recommended to pay moreattention. For example, the predefined code may implicate that an index value of 3-5implies that it requires more attention and intervention is recommended. Forexample, the predefined code may implicate that an index value of 1 -3 implies that 30 intervention is required. In example 2 below, there is shown the predefined PI valuecode table used to determine the PI value. In addition, various clinical tests may beperformed wherein the health care providers/experts are asked periodically to enter (ina blind or non-blind fashion) a PI value, that may be used afterwards for validationand fine-tuning of the algorithms’ decisionmaking. 30

According to additional embodiments, the fuzzy logic inference may be builtusing the consensus of the health care providers’ data, such as may be obtained asexplained above. For example, membership functions may be assigned for each 5 parameter that is chosen to create the index, and the ranges for the various parametersmay be determined. For example, the ranges may include descriptions, such as,Normal, High, Very-High, Low and Very-Low. The index itself may thus have anumber of membership functions that is determined by the scale of the values of theindex. For example, an index, which is in the scale of 1 to 10, may have ten 10 membership functions, one for each index value. Upon obtaining the membershipfunctions, a rule set may be determined to relate the inputs (of the measuredparameters) to the outputs, using the verbal descriptor of the membership functions,and consequently to determine the value of the index. For example, for an index thatis determined by, for example, 5 parameters, if one parameter is “Very High” and an 15 additional parameter is “High” and another parameter is “Normal” and an additionalparameter is “Low” and another parameter is “Normal”, then the Index value is X.The fuzzy logic operators used may include, for example, minimum for AND,maximum for OR, maximum for Aggregation and centeroid (center of gravity) of areafor de-fuzzification. In example 2 hereinbelow, there are demonstrated exemplary 20 fuzzy logic membership functions and an exemplary fuzzy logic rules matrix.

According to additional embodiments, the design of the fuzzy inference of theindex model that is based on several health care providers’ know-how may be furtherenhanced or fine-tuned by comparing to the average, or any other statistical results 25 such as median or mode, index values determined by the health care providers. Thefuzzy logic interface of the index model may further be validated. For example,validating the fuzzy inference model may be performed by an additional questionnaireor on real data, including sample cases. In the additional questionnaire, the Indexvalues determined/calculated by the fuzzy inference model is shown, and the health 30 care provider may opine on the values. By considering the opinion of the health careproviders, fine-tuning of the fuzzy logic model may further be performed.

Additional information regarding fuzzy logic may be found, for example at:“Tutorial on Fuzzy logic”, by Jan Jantzen, 1998, 31 http://fuz2y.iau.dtu.dk/download/logic.pdf, the content of which is incorporated hereinby reference in its entirety; “Fuzzy logic for just plain folks”, chapters 1-3, that maybe found at http://www.fuzzy-logic.com/Chl.htm, http ://www. fuzzy- logic. com/Ch2.htm, http://www.fijzzy-logic.com/Ch3.htm, the contents of which are 5 incorporated herein by reference in their entirety; “A touch of gray” by Tran T. andZomorodi, Z., that may be found at http://www.duke.edu/vertices/win94/fuzlogic.html, the content of which isincorporated herein by reference in its entirety. 10 According to some embodiments, when determining the ranges of the parameters that are used for the calculation of the index, several considerations shouldbe taken into account, such as, for non-limiting examples, the age of the patient,weight of the patient, gender of the patient, current and/or prior medical condition ofthe patient, medications (currently and/or previously administered), known respiratory 15 or cardiac disorders, pacemaker or other implant(s) in the patient, transplanted organs,and the like. Those considerations may be important for a respiratory/pulmonaryindex, since various parameters may have different ranges for different patients. Forexample, it is known that there are differences in the normal ranges of HR and RRbetween adults and children at different ages, and hence, the Pl values for the same 20 parameters may differ for different classes of patients. Thus, ΡΙ-models may bedesigned for distinct ages, to account for the changes in the ranges of normalrespiration rate and heart rate at different ages. Moreover, at ages lower than 1 year,the parameters value depends also on the child’s weight Therefore, various modes ofoperation may be used when determining a PI index, wherein the modes of operation 25 takes into consideration additional parameters (factors) which may include specificcharacteristics of the patients, such as, for example, age, weight, gender, medicalcondition, medication, and the like, or any combination thereof. For example, withregards to age, several groups may be determined, each with its particular normalranges of various parameters, such as HR and RR. For example, for a newborn (age 30 of up to one month), normal RR may be in the range of, for example, about 30 to 60BPM and the normal HR may be in tlie range of, for example, about 100 to 160 beatsper minute. For example, for an infant (age of one month to one year), normal RRmay be in the range of, for example, about 30 to 40 BPM and the normal HR may bein the range of, for example, about 90 to 150 beats per minute. For example, for a 32 child (age of one to two years), normal RR may be in the range of, for example, about22 to 30 BPM and the normal HR may be in the range of, for example, about 80 to125 beats per minute. For example, for a child (age of 3 to 5 years), normal RR maybe in the range of, for example, about 20 to 24 BPM and the normal HR may be in the 5 range of, for example, about 70 to 115 beats per minute. For example, for a child (ageof 6 to 12 years), normal RR may be in the range of, for example, about 16 to 22 BPMand the normal HR may be in the range of, for example, about 60 to 100 beats perminute. For example, for a child (age of 1 to 2 years), normal RR may be in the rangeof, for example, about 22 to 30 BPM and the normal HR may be in the range of, for 10 example, about 70 to 125 beats per minute. For example, for a child (age of 3 to 5years), normal RR may be in the range of, for example, about 19 to 25 BPM and thenormal HR may be in the range of, for example, about 70 to 110 beats per minute. Forexample, for a child (age of 6 to 1 year), normal RR may be in the range of, forexample, about 14 to 24 BPM and the normal HR may be in the range of, for 15 example, about 70 to 110 beats per minute. For example, for an adult (age over 12years), normal RR may be in the range of, for example, about 10 to 18 BPM and thenormal HR may be in the range of, for example, about 60 to 100 beats per minute.Changes to the fuzzy logic interface model used to determine the index may apply tothe membership functions of the parameters whose ranges are influenced by the 20 specific patients parameters. For example, for the age influence on the PI index, thechanges to the fuzzy logic model may be in the ranges of tlie RR and HR membershipfunctions, as further illustrated in Fig. 2, described below.

According to some embodiments, the calculations of the index may use 25 averaged input values, in order to prevent noise. The averaging time window may beconstant (such as, of example, 30 sec for adult mode, 15 sec for pediatric modes). Inaddition, or alternatively, the averaging time window may be a non-constant and/oradaptive time window that be determined by various algorithms. 30 According to other embodiments, several special cases may arise during monitoring that may require specific attention. Such cases may include, for example,a condition wherein no breath is detected (Apnea); an Auto Zero condition; ameasuring probe is not connected, and hence the measured parameter is not recorded.In the following cases the monitoring device may present a PI value: Apnea - once 33

No-breath is detected (more than Xsec with no new detection of CO2, X is defined bythe user, usually 20-30 sec), EtCO2=0, RR-0. These values are valid, and the PI isexpected to slowly decrease. Once breathing resumes, EtCO2 >0 after one breath,RR>0 after 2 breaths, PI is handled the same way, that is, it is expected to slowly 5 increase. Auto Zero - PI may present the last value. When a measuring probe isdisconnected, the averaging may be carried out using the available measurements,disregarding the not-valid measurements. Only after the time window for averagingpasses, would there be a non-valid input and hence the PI would be not-valid as well. 10 According to some embodiments, when determining the value of the index, for example, by using a fuzzy logic interface, non-linear, or any form of interactionbetween various measured/sensed parameters may be taken into consideration whencalculating the index value. Such interaction between the parameters may include, forexample, synergistic effects between various parameters. At some cases, a 15 combination of parameters may have a synergetic effect, and by using fuzzy logicrules, the non-linear interactions between parameters is evident, and thus the synergybetween the input parameters used for calculation of the index may be captured. Forexample, an example of a synergetic effect of the rules of the fuzzy logic interfaceused to determine a PI value: Higher than normal RR level alone could be perceived 20 as a risk, however, when it is accompanied by a higher than normal level of ETCO2,the risk level is lower, as this is the expected normal physiological response. Severalhealth related parameters may be used in the calculations. For example, as illustratedin example 2 hereinbelow, If (EtCO2 is Very High) and (RR is Very High) and (SpO2is Normal) and (HR is High) then (PI is 2)). Such calculation takes into consideration 25 the various membership functions and synergistic effects between various parameters.In this example, when EtCO2 is very high and RR is very high, the PI=3, however,while SpO2 is normal and hence it would not affect the PI value, the HR is High,which leads to a reduction of one level in the PI value, which becomes 2. In adifferent example, if EtCO2 is high and RR is normal and SpO2 is low and HR is 30 normal, then PI-6. In this example, when EtCO2 is high and RR is normal then PI“7,however, since SpO2 is low, it leads to a reduction of 1 in the PI. Since HR is normal,it does not further affect the PI, and hence the PI=6. 34

According to some embodiments, the fuzzy logic interface may further beadjusted for noise and artifacts filtering, wherein noise and artifacts are events that areidentified visually and/or by other identification means, which are not identified asauthentic medical conditions. For example, the inputs to the fuzzy logic (interface) 5 module may be averaged over a moving time window using a constant window length(such as, for example, 30 sec for adult mode, 15 sec for pediatric mode) or an adaptivewindow length. The adaptive window length may be pre-defined, for example, by thevariability of the inputs at the most recent measurements, as may be determined, forexample, by various algorithms; or alternatively, may be dynamically adjusted, as by 10 expanding or contracting the time window in response to values of criteria. Forexample, the inputs to the fuzzy logic may be filtered out from averaging and furthercalculations as a result of identification of artifacts. The artifacts’ identification maybe based, for example, upon CO2 waveform analysis. For example, when a waveformis classified as an artifact (for example, when a patient is eating, talking, and the like), 15 the EtCO2 and RR related to that time period may be updated using a modifiedmathematical algorithm for averaging. Alternatively, that time period may beignored. For example, when a patient is receiving oxygen supply, the oxygen supplymay result in dilution of the CO2 signal. Correction for such dilution artifact may bedone, for example, for the EtCO2 level. 20

According to further embodiments, the change in the PI value over time (PItrend) may be displayed graphically. The graphic display may exhibit the PI trendedover the last “n” (time units) of monitoring. For example, n may be any time periodin the range of 5 minutes to 12 hours. This display may be used to indicate the 25 patient’s status, such as, for example: stable, improving, deteriorating, as well asproviding a depiction of the rate and change of the patient’s status. Displaying of thePI trend may simplify the assessment of the changes in the ventilatory condition of thepatient as compared to assessing the patient condition based on the trend of theindividual parameters When looking at the trends of the individual parameters, it 30 may not be easy and intuitive to determine the patients status and change in status,without taking into consideration tlie absolute values of the individual parameters andtheir interactions, since both an increase or decrease in any of those parameters maybe “good” (improvement) or “bad” (deterioration), depending on the absolute value ofthe parameter. For example, a decrease from a higher than normal absolute value 35 towards the normal absolute value may be considered “good”, while a decrease from anormal value to a lower than normal value may be considered “bad”. Likewise, anincrease from a lower than normal value towards the normal value may be considered“good”, while an increase from a normal value to a higher than normal value may be 5 considered “bad”. The PI trend may be depicted as a graphic display of the PI valuesover time. The duration period of the trend may be chosen to be over any time periodin the range of, for example, between 5 minutes to 12 hours of the last measurements.The resolution of the graphical display may change accordingly in correlation to theselected time period. By providing a trend of the PI index an estimation of how the 10 patient’s respiratory status is changing over time, such as, stable, improving ordeteriorating may be obtained. This may be attributed to the fact that the PI itself isan overall picture of the patient’s respiratory status, and the changes in value of thisindex may provide a clear picture as to whether the patient condition is changing ornot. 15

According to additional embodiments, the trends of the parameters (that is, thechange of the values of the parameters over time) may also be taken intoconsideration when determining the index and may further be displayed. 20 According to further embodiments, an index of reliability may also be determined. The index of reliability (referred to herein also as "reliability index" or"RI") may provide a measure of the reliability of the data and more specifically, thereliability of the PI. For example, the reliability index may be used to predict andanticipate artifacts. The reliability index may be determined, for example, by analysis 25 of the CO2 waveforms, as depicted by a capnogram. If breath flow is also measured,its waveforms may also be used for this purpose. The use of breath flowmeasurements may refine and improve the index of reliability. Breath flowwaveforms strongly complement the waveforms created by the CO2 measurement,since both measurements represent essentially the same event, which is the breath 30 cycle. While the breath flow relates to the envelope of the waveform, the CO2 relatesto the CO2 concentration within the envelope. Using both parameters may reveal anduncover and assist in distinguishing between a valid measurement, a noise, an artifact,and the like. The reliability index may assist the user (the health care provider) todecide if a low PI value or any of the other measured values is real (and represent a 36 genuine clinical event), transient or an artifact. The reliability index may furtherassist the health care provider in assessing how much credibility may be attribute tothe displayed PI value. In addition, the reliability index may allow the health careprovider to detect artifacts, such as when the monitoring device is not placed properly 5 on the patient, the monitoring device is not measuring properly, and the like. Thereliability index may be determined from both analysis of the CO2 waveform andrespiration rate pattern. By obtaining data from controlled studies, the characteristicpatterns attributed to artifacts may be defined. Analysis of the real time waveformdepicted by the monitoring device and comparison to the known artifact patterns may 10 be used to calculate the reliability index.

According to yet further embodiments, confidence (reliability) level mayaccompany the index value. The confidence (reliability) level may further bedisplayed for the health care provider. The confidence (reliability) value may be a 15 number in the range of, for example 0 to 1. The confidence (reliability) level valuemay be determined/calculated from the standard deviation (std) of the various inputvariables (measured/sensed) parameters, as well as artifacts, if such are identified.Furthermore, the various input variables may be weighted according to theft-confidence (reliability) level. Other mathematical, including statistical, methods of 20 calculating or determining confidence intervals levels or other interval estimates maybe used as will be understood by one of skill in the art.

According to additional embodiments, a pause frequency parameter may bedetermined. This parameter may include a measure of events wherein no breathing is 25 detected over a period of time. The events of lack of breathing may include, forexample, pause and apnea events; and the pause frequency parameter may include ameasurement of the patient's pause and apnea events over a period of time. Veryoften patients may stop breathing for short periods of time either because ofmechanical obstructions or sometimes because of a central (brain) block. The pauses 30 (apnea events) are periodic and their frequency may be indicative of the condition ofthe patient. The pause frequency parameter may be determined/calculated from theCO2 waveform, as obtained by the capnogram. A pause event may be defined, forexample, as any inhalation stage that persists for longer than any number of secondsin the range of, for example, 5 to 40 seconds (such as for example 20 seconds), and 37 proceeds after an exhalation period lasting less than any number of seconds in therange of, for example, 5 to 20 seconds (such as, for example, 10 seconds). The timeperiods may be determined, for example, according to the average time of the lastthree exhalation cycles. Such determination of a pause event may be used to exclude a 5 slow, rhythmic breathing pattern from being defined as a group of pause events. Inaddition, a maximum time out of, for example, 100 seconds may be determined. If apause is detected, a new pause can only be counted if at least three new, valid breathcycles were detected beforehand. Thus, the pause frequency parameter may bedefined by the number of pause events per period of time (such as, for example, an 10 hour). The pause frequency may be updated at any time interval, such as for example,every 5 minutes, after the period of 1 hour. The values of the pause frequency mayfurther be stored and used for the display of the pause frequency trend, wherein thetrend data represent the change of the pause frequency over time. According to someexemplary embodiments, during the first hour (when insufficient data has 15 accumulated), a value may be provided and updated, for example, every 15 minutesuntil 1 hour has been reached (wherein during this time period the frequency isdetermined/calculated as if it was determined/calculated for 1 hour). During this timeperiod an indication showing that the pause frequency is still based on a shorter periodthan 1 hour may be displayed. Since the health care provider, such as a nurse, may 20 not be constantly present next to the patient and/or the monitoring device, and maynot constantly track (monitor) the patient condition, a parameter, such as the pausefrequency parameter, which is a periodic type effect may not easily be observed bythe health care provider if not otherwise tracked by the monitoring device. Inaddition, according to further embodiments, the pause amplitude parameter may also 25 be determined. The pause amplitude parameter may include the time length (such as,for example, in the range of 5 to 60 seconds) of each of the detected pause events andthe dispersion of the time length of those pause events over a period of time (such as,for example, over a time period of 60 minutes). 30 According to some embodiments, various additional parameters (factors), in addition to the main parameters according to which the index is determined, may beadded for the calculations of determining the index. The additional parameters(factors) may include various patient specific parameters that may be characteristic ofthe patient. The additional parameters (factors) may include “on-line” parameters that 38 are health related parameters that may be sensed/measured on-line, in real-time,and/or may include other input(s) (such as, for example, by a keyboard, switches ortouch panel) or other parameters (such as, for example, age, gender, weight, and thelike). The additional parameters may include “off-line” parameters that may include 5 parameters that are related to the patient’s medical history and demographic and/orparameters that relate to a recent/current health condition of the patient. In addition,the trends of the additional parameters (that is, the change of the values of theparameters over time) may also be taken into consideration when determining theindex. Using additional parameters for calculation of the index may be dynamic, that 10 is, according to necessity or variability or of averaging weight(s), at any given time.The additional parameters may be added either manually or by any route ofcommunication. For example, with respect to a PI index that is determined by theparameters of HR, RR, EtCO2 and SpC>2, additional on-line parameters may includesuch parameters as, but not limited to: CO2 wave form and CO2 wave form related 15 parameters, such as, for example, changes in ETCO2, a slope of the increase in theCO2 concentration, a change in a slope of the increase in the CO2 concentration, timeto rise to a predetermined percentage of a maximum value of CO2 concentration, anangle of rise to a predetermined percentage of a maximum value of CO2concentration, breath to breath correlation, CAP-FEV1 (forced expiratory volume 20 over 1 sec obtained from at least one capnographic measurement, a measure of flow),CAP-FEV1/FVC, inhalation to exhalation ratio (for example, the inhalation toexhalation ratio, multiplied by the EtCC>2 measured values, may be used to determinehow much CO2 is being ventilated), duty cycle of the CO2 (which is related tomeasure of minute ventilation), and the like; breathing related parameters; heart 25 function related parameters, such as, for example, blood pressure, NI blood pressure,systolic to diastolic ratio, and the like; neurological parameters; systemic perfusionrelated parameters; visual parameters; flow rate; spirometry; and the like, or anycombination thereof. For example, additional off line parameters may include suchparameters as, but not limited to: parameters that are related to a patient’s history and 30 demographic such as, for example, age, gender, weight, current diagnosis, medicalhistory (for example: smoking, heart disease, lung disease, sleep apnea, medicationstaken, pace maker) and the like, and any combination thereof. Additional off lineparameters that are related to a recent health condition of the patient may include suchparameters as, but not limited to: medical treatments (for example medications given, 39 ventilation, oxygen supply), lab tests (for example, blood gases, pH, general bloodtests, urine tests), whether the patient is intubated or not, whether the patient is asleep(during sleep, stable CO2 waveforms are expected, non-stable waveforms areindicative of poor condition, while in an awake patient the opposite is expected, that 5 is, non-stable waveform are normal, while stable wave forms may be indicative of apoor condition); and the like, or any combination thereof.

According to some embodiments, the additional parameters (factors) may beused to manipulate the inputs to the membership functions, for example, by changing 10 the limit ranges of various measured parameters. For example, evaluation of HeartRate (HR) ranges of normal, high and low may further depend on the patient’s age.Accordingly, the parameters of the membership functions may be defined as a linearor non-linear function of age. Reference is now made to Fig. 2, which illustrates anexemplary value of membership function as a function of age. The graphs shown in 15 Fig. 2 represent the value of the membership function (mf) of the heart rate (HR,beat/pulse per minute (bpm) as a function of Age (years). As shown in Fig. 2, themembership function generates a trapeze shape, when using 4 parameters. When theparameters depend on age, different shape may be generated, according to the age.For example, at the age of 1 year old, the range of normal heart rate is at higher bpm 20 than the range for normal heart rate for the ages 12-60 years old (the parameters forage 1 are: 70 80 135 145 bpm, and for ages 12-60: 40 60 100 120 bpm). Followingthe parameters that define the boundaries of the trapezes (bold lines) and the trapezes(thin lines) illustrates the dependency of the membership function on age and HR. (Inthe graph, The X- axis is age in years; the Y-axis is Heart rate in bpm; and the Z-axis 25 is the membership function value for Normal HR in a scale 0-1.

According to further embodiments, the additional parameters (factors) may beused for the creation of additional rules that may be implemented with, for example,the fuzzy logic interface model. The additional parameters may be used as any of tiie

30 basic (primary) parameters, using their membership functions and incorporation intheir rules, such as exemplified in Fig 4A, belowherein. The additional rules may bein a flat or hierarchical structure. For example, an additional on-line parameter thatmay be used for the calculation of a PI index may include a parameter such as, forexample, the diastolic blood pressure (BP), that may be used in conjunction with HR 40 as shown, for example in the following rules: If HR is High and BP is High thenPI=PI (normal HR) -2; If HR is High or Low and BP is Normal PI=PI (normal HR)-1;and the like. 5 According to further embodiments, the additional parameters (factors) may be used for the creation of an adaptive index risk function. The value of the indexreflects the risk to a patient’s health. The risk association with a specific index valuemay differ at different conditions. Such different conditions may allow the use ofadditional parameters such as, for example, the patient’s illness, the hospitalization 10 department, and the use of such additional parameters may be used to revise the riskassociation function. For example, with respect to the PI index, the value of the indexreflects the risk to a patient’s health, such that, for example a PI index value of PI"[8-10] means that the patient is in a good condition. The risk association with a specificindex value may differ at different conditions. Thus, a linear Pi-Risk function may be 15 altered to reflect higher sensitivity for changes at low PI values, or at high PI values.The PI may be determined/calculated as is, and its final value may be changedaccording to the association functions. Such risk association function is illustrated,for example, in Fig, 3, which illustrates graphs depicting the PI value (on a scale of 1to 10, Y-Axis) with relation to the level of risk (from low to high, x-axis). Line a in 20 the graph illustrates a linear Pi-Risk function; line b illustrates a graph which reflecthigher sensitivity for changes at low PI values; line c illustrates a graph which reflecthigher sensitivity for changes at high PI values. In this example, PI, which equals to 5in line a, may turn to 2 using line c association rule. 25 Reference is now made to Fig. 4A, which illustrates an exemplary block diagram scheme of a method of calculating a PI index using a fuzzy logic interface,according to some embodiments. As shown in Fig. 4A, input main parameter, suchas parameter 170A, enters a characterization process (such as process 172A). Theinput parameters may include any number of sensed/measured/determined patient 30 parameter that may be directly or indirectly sensed/measured/determined. The inputparameters may include any plurality of sensed/measured/determined patientparameters. For example, the input main parameter may include such parameters as,but not limited to: heart rate, respiration rate, CO2 related parameters, O2 relatedparameters, electrocardiogram (ECG), encephalogram (EEG), blood pressure, 41 spirometry, and the like, or any combination thereof. In the characterization process,the respective membership value to each characteristic group (such as groups 173 A-C) is calculated/computed/assigned. The characteristic groups may be labeled, forexample, as: “Normal”, “High”, “Low”, “Very High”, “Very Low”, and the like. The 5 same process may be applied onto any number of input main parameters, such as, forexample, input parameter 170B, which may be characterized by characterizationprocess 172B, where the respective membership value to each characteristic group(such as groups 173D-F) is calculated. At the next step, various combinations of pairsof each group enter an Output Characteristic Group (such as, for example, output 10 characteristic groups 174A-n, wherein n may include any integer number higher than0). For example, as shown in Fig. 4A, characteristic groups 173A and 173D mayenter Output Characteristic Group 174A. The membership values of both parametersgroups, the weights related to each and a logic operator combining them (such as, forexample, “AND”, “OR”) may be used to calculate/determine/compute/assign the 15 membership value for each Output Characteristic Group. The topology of the network is designed, for example, using medical experts’ knowledge, clinical trials, literature,sample cases, and the like. Eventually, an output value (such as output value 175) iscalculated from all values of membership functions, for example, by an aggregationfunction (such as, for example, MAX, and the like) and/or by an averaging function 20 (such as, for example, centeroid, mean, average, and the like). According toadditional embodiments, the method of calculating an index may further include theuse of various additional parameters (factors), in addition to the main parameters. Theadditional parameters (factors) may include various patient specific parameters thatmay be characteristic of the patient, such as, for example, blood pressure, age, 25 medical condition, sedation, patient awake, patient asleep, and the like. For example,an additional parameter (factor), such as, for example, additional parameter (factor)176 may enter into one or more of the characterization processes and may thuscontrol/tune the membership functions and membership value for each OutputCharacteristic Group that result from the characterization process. 30

According to additional embodiments, the calculation model may include a

Bayesian network. A Bayesian network, as known in the art, may include a graphical

representation of relationships between variables based on probability theory. A

Bayesian network includes a directed a-cyclic graph with nodes representing variables 42 that are associated with a table of conditional probabilities. Nodes are connected byarcs, which model the relationships between variables. The probability of a node canbe calculated when the values of other variables are known. A Bayesian network has alearning capability using known cases and it may tolerate missing data. Topology and 5 probabilities of a Bayesian network may be assigned using, for example, experts’knowledge, textbooks, clinical trials, sample cases, and the like. Bayesian networknodes may include various parameters and/or inputs and/or factors, that may include,for example: measured/sensed parameters (such as; for example, EtCO2, RR, SpO2,HR, blood pressure, and the like); medical condition (such as, for example, heart 10 failure, asthma, smoking, lung disease, and the like); demographics (such as, forexample, age, gender, weight, and the like); lab test results (such as, for example,blood gases, pH, urine tests, and the like); medications (such as, for example, steroids,sedatives, anti-inflammatory drugs, and the like); and the like. Hie node may havediscrete or continuous values. Considering the various parameters and/or inputs, a 15 Bayesian network may be used to compute/evaluate/determine a health related index,such as, for example, a PI that represent the probabilities of the status of the patient.The Bayesian network may further be used to compute/evaluate/determine theprobabilities of a diagnosis (such as, for example, hyperventilation, hypoventilation)and may further provide recommendation for additional tests and/or interventions. 20 Additional information regarding a Bayesian network may be found, for example at:"A Tutorial on Learning with Bayesian Networks” by Heckerman D, in “Learning inGraphical Models” by M. Jordan, ed. MIT Press, Cambridge, MA, 1999, the contentof which is incorporated herein by reference in its entirety; "Bayesian Networks” byBen-Gal in F. Ruggeri, R. Kenett, and F. Faltin (editors), Encyclopedia of Statistics in 25 Quality and Reliability, John Wiley &amp;amp; Sons (2007), the content of which is* incorporated herein by reference in its entirety. Reference is now made to Fig. 4B,which illustrates an exemplary Bayesian network topography, according to someembodiments. As shown in Fig. 4B, exemplary Bayesian network 120, may illustraterelationships (represented by arrows) between various parameters and/or factors 30 and/or inputs, such as, for example, Blood Pressure (121), Age (123), Heart rate (122)SpO2 (126), Respiration Rate (124), EtCO2 (125). By assigning the variousrelationships between the various factors an output (127), such as an index, (such asPI) may be determined/calculated/computed. 43

According to additional embodiments, the calculation model may include adecision tree. A decision tree is a predictive model that may be used to mapobservations regarding an item to conclusions regarding the item's target value. Thetree has a hierarchical graph structure consisting of nodes and directed edges. The top 5 layer may include input nodes (such as, for example, parameters and/or inputs/ and/orfactors, such as, but not limited to: measured/sensed parameters (such as, for example,EtCO2, RR, SpO2, HR, blood pressure, and the like); medical condition (such as, forexample, heart failure, asthma, smoking, lung disease, and the like); demographics(such as, for example, age, gender, weight, and the like); lab test results (such as, for 10 example, blood gases, pH, urine tests, and the like); medications (such as, forexample, steroids, sedatives, anti-inflammatory drugs, and the like); and the like).Decision nodes determine the order of progression through the graph. The “leaves” ofthe tree are all possible outcomes or classifications, while the “root” is the finaloutcome, that may include, for example, the patient status classification, 15 recommendations for intervention, and the like. The tree may be built using experts’knowledge, sample cases, literature, clinical trials, and the like. Methods for buildingdecision tree may include, for example, algorithms, such as, C5.0, CART and ID3which are decision tree algorithms well known in the art. There may be reinforcementlearning to refine the tree structure and parameters. The outputs classification may be 20 assigned with corresponding probabilities. An ensemble of trees may be generated toincorporate different trees, to produce a more complex classification. Additionalinformation regarding decision trees may be found, for example at: " Improved Use ofContinuous Attributes in C4.5”, by J. R. Quinlan, Journal of Artificial IntelligenceResearch, 4:77-90, 1996, the content of which is incorporated herein by reference in 25 its entirety. Reference is now made to Fig, 4C, which illustrates an exemplarydecision tree graph structure, according to some embodiments. As shown in Fig. 4C,exemplary decision tree graph structure 140 may illustrate directed edges and nodes(represented by circles and arrows), which relate between various parameters and/orfactors and/or inputs, such as, for example, age, Coronary heart disease (CHF), heart 30 rate (HR), blood pressure and their predicted outcome. The nodes may include inputnodes and/or decision nodes, which may be used to determine the order of progressionthrough the decision tree graph. For example, as illustrated in Fig. 4C, input node 141(age) may lead to two decision nodes, represented by arrows, for example, age isbelow 60 or above 60. According to each of the decision nodes, progression to other 44 input nodes is determined. For example, as shown in Fig. 4C, if age is above 60, CHFinput node (142) is considered. If the CHF input node is positive, then another inputnode, such as, for example, blood pressure (144) is considered. In addition oralternatively, in parallel or in series, other input nodes may be considered. As shown 5 in Fig. 4C, Heart rate (HR) input node (143) may be also considered in parallel. If theHR is high, then the blood pressure input mode (144) is also considered. Additionaldecisions nodes are represented by tripled dots. The “root” of the tree represents thefinal outcome of the decision tree. For example, the outcome may include if thepatient status (condition) is the same (such as, for example, outcome 146), or if the 10 patient status (condition) is worse (such as, for example, outcome 145).

According to additional embodiments, the calculation model may include afeed forward neural network. A feed forward neural network is a biologicallyinspired classification algorithm. It may include simple neuron-like processing units, 15 organized in layers. Every unit in a layer is connected with the units in the previouslayer. The connections may not all be equal, as each connection may have a differentstrength or weight. The weights of the connections encode the knowledge of anetwork. Often the units in a neural network are also called nodes. Data enters thenetwork at the inputs and passes through the network, layer by layer, until it arrives at 20 the outputs. During normal operation, that is, when it acts as a classifier, there is nofeedback between layers. During the learning phase, the weights in the network maybe modified. All weights are modified in such a way that, when a pattern is presented,the output unit with the correct category will have the largest output value. Learningmay be carried out using the back-propagation algorithm. Reinforcement learning is 25 also possible. A hierarchical structure of networks may also be designed. The inputsto the network may include, for example, measured/sensed parameters (such as, forexample, EtCCb, RR, SpO2, HR, blood pressure, and the like) and/or variousadditional parameters (factors) that are patient specific and may be characteristic ofthe patient, such as, for example, blood pressure, age, medical condition, sedation, 30 patient awake, patient asleep, and the like. The output may include, for example, ahealth related index, such as, a PI that may represent the patient status classification.The network may also capture changes over time (trend) of the input parameters byadding input nodes of previous time points. Additional information regarding feed-forward neural networks may be found, for example at: “Pattern classification” (2nd 45 edition), by Duda, R.O., Hart, P.E., Stork, D.G., (2001) Wiley Publishers, the contentof which is incorporated herein by reference in its entirety; “Neural Networks forPattern Recognition” by Bishop, C.M. (1995), Oxford University Press. 5 Reference is now made to Fig. 4D, which illustrates an exemplary feed forward neural network, according to some embodiments. As shown in Fig. 4D,exemplary feed forward neural network 160, may illustrate propagation ofinformation (represented by arrows) from various input measured/sensed parameters,such as, for example, Heart rate (HR, 161A), Respiration rate (RR, 161B), SpO2 10 (161C), EtCO2 (161D), to and through neuron-like processing units (layers) of the network (such as, for example, hidden layers 162A, 162B and 162C). From theinformation processed/evaluated/determined in the various layers of the network, anoutput (164 in Fig. 4D), such as, for example, PI, may be determined. The weightsand relationships between the various layers may be modified, for example, by 15 employing a feedback mechanism, which may be used, in particular, during a learningphase.

According to further embodiments, when determining the index value, ateaming model may be used in order to define or refine the index risk association 20 function, and/or to fine-tune rules and/or to fine-tune membership functions,according which the index is determined/calculated. As shown in the Fig. 5, alearning model scheme (100) may include three main blocks: 1. Environment (102),which sends measured parameters as inputs to the model. The parameters may includesuch parameters as, but not limited to: ETCO2, SpO2, RR, HR, and the like. 2. The 25 calculation model (104) may include various mathematical models, such as, forexample, but not limited to: Fuzzy inference model, Bayesian network, decision tree,neural networks, radial base functions, linear regression models, and non-linearregression models. For example, the calculation model may include a fuzzy inferencemodel. The calculation model may then calculate an output (106) that may include, 30 for example, the PI value. For example, the calculation model may include aBayesian network. For example, the calculation model may include a decision tree.For example, the calculation model may include a feed forward neural network. 3.Performance criteria (108) may include a score that evaluates the reliability (quality)of the prediction of the model. The criteria may be estimated by, for example, a 46 health care provider expert. In addition, or alternatively, the criteria may be anautomatic estimation using additional data collected regarding patient status andmedical interventions. Using the performance criteria, a learning process (110) maybe applied in order to maximize the performance until convergence is attained. 5 Learning and optimization may be carried out using various methods, such as, forexample, but not limited to: neural networks, a support vector machine (SVM),genetic algorithms, simulated annealing and expectation-maximization (EM), and thelike. Learning may be carried out off-line to gain a better general model, or on-line asan adaptation of the model to fit each specific/individual patient. In the last case, the 10 longer the patient is monitored, the more over-fitting of the model towards his state.

According to some embodiments, the monitoring device may further includeand display, medical recommendations to the health care professionals. The medicalrecommendations may be deduced from analysis of at least some of the individual 15 parameter’s values and patterns and comparison of the measured values and patternsto the known ranges and patterns of the individual parameters. Theserecommendations may be displayed, in addition to the indications derived from thedetermined/calculated condition-index-value. The medical recommendations may bebased on, for example, the characteristics of the CO2 waveforms (presented as a 20 capnogram) and RI values. The recommendation provided to the health care providermay include, for example such recommendation as: If CO2 waveforms are observedwith characteristics indicative of a partial obstruction (such as long downward slopeof the waveform), then the device may recommend “open airway” or “check airway”.If the CO2 waveforms are very low but exhibit an excellent form indicative of low 25 blood flow to lungs, then the device monitor may recommend: “check blood pressure”or similar. The detection of known CO2 waveform patterns, which are indicative ofknown patient conditions that should and could be treated to improve the patient care,may be used in triggering and issuing the recommendations to the health careprofessional. For example, it is very common that a patient entering partial 30 obstruction promote a waveform pattern wherein the downward slope of thecapnogram increases in time, with a gradual fall. In such an instance, a notification tothe health care provider, such as “check patient airway” may be issued. For example,if the CO2 waveforms pattern become very rounded and low, which is indicative of amechanical problem with the cannula of the capnograph, a recommendation, such as, 47 “check cannula interface” may be issued. Those and other similar medicalrecommendations issued by the monitoring device may be presented in addition to theindications derived from the PI. Examples of such analysis of waveform pattern maybe found in publication: Krause B. and Hess D.R. (2007) “Capnography for 5 procedural sedation and analgesia in the emergency department”, Ann Emerg Med.50(2), Pages 172-81, incorporated herein by reference. For example, lower airwayobstruction may be seen by a capnogram showing a curved ascending phase and up-sloping alveolar plateau that indicate the presence of acute bronchospasm orobstructive lung disease; for example, distinguishing between bradypneic 10 hypoventilation from hyperventilation may be performed according to the CO2 waveform: while in the bradypneic hypoventilation decreased respiratory rate, highamplitude and wide capnogram are detected, in hyperventilation an increase inrespiratory rate, low amplitude and narrow capnogram are detected. 15 According to some embodiments, the medical device may further include a user interface that may allow the user to select the data to be displayed and to controlvarious operating parameters. Moreover, different displays may be included toaccommodate different needs of the different users (such as a nurse, a physician, ananesthesiologist, and the like). Allowing the user to change the view of the data may 20 permit the user to toggle through the different levels of information for furtherevaluation of a condition. For example, the basic screen may display the condition-index-value and the condition-index-value trend data. Changing to the next displaymay reveal the actual (measured) data values and the trends of the values that relate tothe parameters from which the condition-index-value is determined/calculated. 25 Further toggling the display may provide the pause frequency and other relatedanalysis and calculation. The use of the various displays may also allow the user tofocus on the parameters that caused an indication of an event and/or recommendationto the user. 30 According to further embodiments, the user interface may also allow the user to enter information that is characteristic for each patient. The use of characteristicpatient information is necessary to allow accuracy of the various measurements andcalculations. Such information may include, for example, age, weight, height, sex,and the like, of the specific patient, as well as other patient related information, such 48 as, intubation (if tlie patient is intubated or not). For example, the patient size and agemay change the PI calculations: for an adult a respiration rate of 12 BPM would benormal, but 36BPM would be considered high, whereas for a child, respiration rate of12 may be considered low, while 36BPM may be considered normal. In addition, 5 classification detection means of various patients may be utilized, wherein theclassification may be based on parameters such as, for example, age group, weightgroup, sex, intubation, and the like. Using such classification may allow themonitoring device to correct its settings to be appropriate for that relevant patient typeand environment. In addition, the user interface may allow automatic detection of the 10 type of the patient according to the measuring interface being used. For example, atubing interface used to measure CO2 in breath may be different between an adult anda child and accordingly, the user interface may automatically adjust the patientsettings to match the patient size group. 15 Reference is now made to Fig. 6, which illustrates exemplary graphical user interfaces, according to some embodiments. As shown in Fig. 6A, user interfacedisplay 200 may display one or more elements, which are related to thedetermined/calculated index. The content of what is to be displayed may bedetermined by a user. For example, display 200, may include element 202, which 20 may allow the user to select a designated display and to further indicate which displayis selected. Element 204 may display the value of the determined/calculated PI.Element 206 may display the trend (change over time) of the determined/calculatedPI. Element 208 may include graphical indications as to the changes of the value ofthe PI. Element 210 may illustrate the values of one or more of the parameters that 25 are used to determine the index value. Element 212 may illustrate the trend of one ormore of the parameters that are used to determine the index value. Element 214 maybe used to present medical recommendation to a health care provider. The display ofthe various elements may include any type of display, such as, for example, anumerical value display, a graph display, a chart, a table, a graphical indication 30 display, a colored-index display, and the like. Reference is now made to Fig. 6B,which illustrates an exemplary graphical display of a PI index, according to someembodiments. As shown in Fig. 6B, graphical user interface, 250, may displayvarious elements. For example, element 252 may allow the user to select a designateddisplay and to further indicate which display is selected. For example, the various 49 WO 2U09/063446 displays may include, trend, system, patient type, alarms, print, zoom, and the like.Element 254 may include a numerical display of the PI value. Element 258 mayinclude a graphical indication (such as, for example, an arrow) of the change in the PI.Element 256 may include a display of the trend of the index over a selected period of 5 time, such as for example, 2 hours. The trend display may be, for example, in theform of a graph which depicts the changes in the PI value over time. Elements 260A-D may display values of various parameters where, according to at least some ofthem, the PI is determined/calculated. For example, as shown in Fig. 6B, element260A displays the level of EtCO2 (in units of mmHg) as a numerical value. For 10 example, element 260B illustrates the respiratory rate (RR, in units of respirations per .minute) as a numerical value. For example, element 260C illustrates the SpO2 levels(% saturation) as a numerical value. For example, element 260D illustrates the Heartrate (HR, units of pulses per minute) as a numerical value. In addition, elements 262A-D are graphical indicators of the changes of the various parameters. For example, 15 elements 262A-D may be in the form of an up-facing and/or down facing arrow.Element 264 may further illustrate a graph depicting CO2 waveform (in units ofmmHg). Reference is now made to Fig. 6C, which illustrates an exemplary graphicaldisplay of a PI index, according to some embodiments. As shown in Fig. 6C,graphical user interface 300 may display the trend (change over time) of various 20 elements, such as, for example, a PI value, various parameters used to determine theindex value, and the like. The selected time period over which the trend is observedmay include any time period, such as, for example, 0.1 to 24 hours. The time periodmay be selected by a user and/or may be automatically determined. For example, thetime period for displaying the trend may be 6 hours. For example, element 304 may 25 include a numerical display of the PI value. For example, element 305 may include agraph display of the trend of the PI value over a selected time period. For example,element 306 may include a numerical display of the heart rate parameter. Forexample, element 307 may include a graph display of the trend of the heart rateparameter value over a selected time period. For example, element 308 may include a 30 numerical display of the SpO2 parameter. For example, element 309 may include agraph display of the trend of the SpO2 parameter value over a selected time period.For example, element 310 may include a numerical display of the respiratory rateparameter. For example, element 311 may include a graph display of the trend of therespiratory rate parameter value over a selected time period. For example, element 50 312 may include a numerical display of the EtCO2 parameter. For example, element 313 may include a graph display of the trend of the EtCO2 parameter value over aselected time period. According to some embodiments, regions of each individualtrend may be shaded, so that the deviations from “normal” regions of the trend are 5 easily distinguished. In addition, or alternatively, the graph may change in color (forexample to yellow and/or red) for those regions that are out of the normal region.Reference is now made to Fig. 6D, which illustrates an exemplary graphical display ofa PI index, according to some embodiments. As shown in Fig. 6D, graphical userinterface, 350, may display the frequency of events wherein deviations from alarm 10 limits (threshold) have been detected over a period of time. Displaying the frequencyof events may include numerical values, graph display, color-coded display, and thelike, or any combination thereof. For example, element 352 illustrates a bar graphshowing the number of times where the respiratory rate was above a high alarm limit(top panel, 353A) or below a low alarm limit (bottom panel, 353B). For example, 15 element 354 illustrates a bar graph showing the number of times where the SpO2value was above a high alarm limit (top panel, 355A) or below a low alarm limit(bottom panel, 355B). For example, element 356 illustrates a bar graph showing thenumber of times where the EtCO2 value was above a high alarm limit (top panel,357A) or below a low alarm limit (bottom panel, 357B), For example, element 358 20 illustrates a bar graph showing the number of times where the heart rate was below ahigh alarm limit (top panel, 359A). For example, element 360 illustrates a bar graphshowing the number of times where the PI value was below a low alarm limit (toppanel, 361A). The time period over which the frequency of events is determined maybe automatically determined and/or may be determined by the user. In addition to 25 displaying the frequency of events, graphical user interface 350 may further displaygraph and/or numerical display of any parameter and/or value. For example, element362 may display the instantaneous measured value of EtCO2. For example, element364 may display the instantaneous measured value of the respiration rate. Forexample, element 368 may display the instantaneous measured value of the SpO2. For 30 example, element 370 may display the instantaneous measured value of the heart rate.

According to some embodiments, there is thus provided a medical device (andsystem) that may be used to monitor a patient's health condition, such as respiratoryand/or pulmonary and/or cardiac status. The device may include, for example a 51 capnography device that may be adapted to sense and/or obtain measurements ofvarious parameters in addition to CO2. Such parameters may include, for example, O2levels, O2 partial pressure, such as SpO2, heart rate, blood pressure, and the like. Thedevice may further include a processing logic that may be used to receive information 5 from at least one of the sensors and to compute/determine/generate a condition-index-value that is directly related to a condition of the patient The processing logic mayinclude any type of hardware and/or software, such as, for example, a processor. Thecondition index value may be in the range of 1 to 10, wherein 10 indicates the bestcondition while 1 indicates the worst condition. The monitoring device may further 10 include one or more display that may be used to present the data collected anddetermined/calculated by the monitoring device. The display may present, forexample, the determined/calculated condition index in numerical format and inindexed format, wherein different ranges along the 1-10 scale of thedetermined/calculated index may be assigned different colors; the change (trend) of 15 the determined/calculated condition index value over time; the reliability of thedetermined/calculated index value; values and patterns of the various parametersmeasured by the various sensors of the monitoring device; graphical indicationsregarding the status of the patient, such as, for example, downward and upwardarrows; and the like. The monitoring device may further be adapted to issue medical 20 recommendations based upon the determined/calculated condition index value andother measured parameters. In addition, the monitoring device may include a userinterface that may allow the user to input patient related data that is specific for thepatient, such as, for example, age, sex, and/or size of the patient. The user interfacemay further allow the user to choose the parameters to be displayed and the form in 25 which the parameters may be displayed, such as, for example, in the form of graphs,numerical values, indicators, and the like.

According to further embodiments, the PI may be used to interface withvarious medical devices and at least partially control the operation of the various 30 medical devices. The PI value may thus be used as a feedback input signal to themedical device, and according to the signal, the operation of the medical device maybe dynamically adjusted, such that the operation of the medical device is better fittedto the condition of the patient. For example, the medical device may include suchdevices as, but not limited to: a patient controlled analgesia (PCA) pump, a dose 52

management medical device, an artificial ventilator, an invasive ventilator, a non-invasive ventilator, and the like. For example, the medical device may include a PCApump that may be used by a patient to self-control the amount of analgesiaadministrated to the patient. When interfacing with the PI, the operation of the PCA 5 may be dynamically adjusted (managed), such that the condition of the patient istaken into consideration. For example, at least one parameter related to the PCA maybe adjusted. For example, when the condition of the patient is deteriorating (asdetermined by the PI value), the operation of the PCA may be adjusted such that, forexample, a lower dose of anesthetic is released with each operation of the PCA. For 10 example, when the patient condition is deteriorating (that is the PI is decreasing), theoperation of the PCA may be adjusted such that the PCA does not release any amountof analgesic. Alternatively, another pump may be activated to provide othermaterials, such as saline, plasma, and the like. The use of the PI to control/adjustoperation of medical devices, such as, for example, a PCA pump, may aid in 15 improving the reliability of that control, since it is based on the PI, which is indicativeof the medical condition of the patient. Additionally, one or more interfaces (such asaudio alarm or flushing light) may be activated independently or in relation to thecontrol/adjustment of medical device operation based on a PI value or change of PIvalue. 20

According to further embodiments, the medical monitoring device may beinterfaced (associated with) one or more medical devices, wherein the operation of themedical devices may be at least partially controlled/adjusted by the medicalmonitoring device. For example, at least one parameter related to the additional 25 medical device may be adjusted/controlled. Controlling the operation of theadditional medical device(s) by the medical monitoring device may be based on aninput, such as, for example, the PI value. For example, the medical devices, which areinterfaced with the medical monitoring devices may include such devices as, but notlimited to: a patient controlled analgesia (PCA) pump, a dose management medical 30 device, an artificial ventilator, an invasive ventilator, a non-invasive ventilator, andthe like.

Reference is now made to Fig. 7, which illustrates a block diagram of amedical monitoring system, according to some embodiments. As shown in Fig. 7, 53 medical monitoring system 400 may include one or more sensors, such as, forexample, sensors 402A-D that may be adapted to obtain/sense/measure various healthrelated parameters. The sensors may include such sensors as, for example, but notlimited to: capnograph, oximeter, spirometer, heart rate sensors, blood pressure 5 sensors, ECG, EEG, Ultrasound, and the like. The parameters thus measured mayinclude, for example, such parameters as, but not limited to: EtCO2, CO2 levels, SpO2,heart rate, blood pressure, flow, CO2 waveform pattern, blood gases, and the like.Medical monitoring system 400 may further include a processing logic, such as, forexample, processing logic 404, that may be used to receive information from at least 10 one of the sensors and to compute/determine/generate a condition-index value that isdirectly related to a condition of the patient. For example, the condition index valuemay be a Pulmonary/Respiratory index. The processing logic may include any typeof hardware and/or software, such as, for example, a processor. The connectionbetween the processing logic and the sensor(s) may include any type of 15 communication route, such as, for example, use of wires, cables, wireless, and thelike. The medical monitoring system may further include one or more displays (suchas, for example, display 406 in Fig. 7) that may be used to present the data collectedand determined/calculated/computed by the processing logic, such as, for example,the index value, the index trend, the parameters used to determine the index, and the 20 like. In addition, the medical monitoring system may further include a user interface,such as user interface 408 that may allow the user to input patient related data that isspecific for the patient, such as, for example, age, gender, weight, height, medicalhistory, current medical status, and the like, or any combination thereof. The userinterface may further allow the user to choose the parameters to be displayed and the 25 form in which the parameters may be displayed.

Reference is now made to Fig. 8, which illustrates a schematic illustration of asystem, according to some embodiments. As shown in Fig. 8, a system, such asmedical monitoring system 500, may include one or more sensors, such as, for 30 example, sensors 502A-D that may be adapted to obtain/sense/measure various healthrelated parameters. The one or more sensors may be connected directly or indirectlyto a patient (such as patient 520). The sensors may include such sensors as, forexample, but not limited to: capnograph, oximeter, spirometer, heart rate sensors,blood pressure sensors, ECG, EEG, Ultrasound, and the like. The parameters thus 54 measured may include, for example, such parameters as, but not limited to: EtCO2,CO2 levels, SpO2, heart rate, blood pressure, flow, CO2 waveform pattern, bloodgases, and the like. System 500 may further include a processing logic, such as, forexample, processing logic 504, that may be used to receive information from at least 5 one of the sensors and to compute/determine/generate a condition-index value that isdirectly related to a condition of the patient. For example, the condition index valuemay be a Pulmonary/Respiratory index. The processing logic may include any typeof hardware and/or software. The connection between the processing logic and thesensor(s) may include any type of communication route, such as, for example, use of 10 wires, cables, wireless, and the like. The medical monitoring system may furtherinclude one or displays (such as, for example, display 506 in Fig, 8) that may be usedto present the data collected and determined by the processing logic, such as, forexample, the index value, the index trend, the parameters used to determine the index,and the like. In addition, the medical monitoring system may further include a user 15 interface, such as, user interface 508 that may allow the user to input patient relateddata that is specific for the patient, such as, for example, age, gender, weight, height,medical history, current medical status, and the like, or any combination thereof. Theuser interface may further allow the user to choose the parameters to be displayed andthe form in which the parameters may be displayed. 20

While a number of exemplary aspects and embodiments have been discussedabove, those of skill in the art will recognize certain modifications, permutations,additions and sub-combinations thereof. It is therefore intended that the followingappended claims and claims hereafter introduced be interpreted to include all such 25 modifications, permutations, additions and sub-combinations as are within their truespirit and scope. 55

EXAMPLES

Example 1 - Calculating PI value using mathematical methods 5 As detailed hereinabove, any of the parameter values used for the calculation of the PIis based on the determined/calculated/computed value of the average EtCO2, averagerespiratory rate, average heart rate and average SpO2.

EtCO2 - End tidal CO2 is measured in units of mmHg RR - Respiration Rate is measured in number of breaths per minute 10 SpO2 - is measured in percentile HR - Heart Rate is measured in pulses per minute

The examples below illustrates index value at various conditions. PI >9-10 15

When the following conditions are realized, the index will be 10 or 9 and no arrowsindicative of hyper or hypoventilation, are displayed:

The following conditions are interpreted as follows: If the respiration rate is in therange of 12 to 28 BPM, and End tidal CO2 is equal to or higher than 28 and smaller 20 that 44 mmHg and SpO2 is higher that 94%, then the determined/calculated/computedindex is 10. If the Respiration rate is in the range of 12 to 28 BPM and End tidal CO2is equal to or above 28 and lower than 44 and SpO2 is in the range of 90 to 94%, thenthe determined/calculated/computed index is 9. Those conditions may be summarizedas follows: 25

If RR is >12 &amp; < 28And EtCO2>28&amp;<44

AndSpO2>94% Then: Index = 10 30

IfRRis>12 &amp; <28And EtCO2>28&amp;<44

And SpO2 > 90% &amp; < 94% Then: Index = 9 35 PI <7 and indication of hypoventilation

IfRRis&lt;5 40

IfRRis<8And EtCO2>64Or EtCO2<12Or SpO2 < 86% 45

IfRRis<5And EtCO2>64OrEtCO2<12And SpO2 < 86%

Then: Index = 4 down

Then: Index = 3 down

Then: Index = 1 down 56

IfRR is &lt;5

And SpO2 < 86% Then: 5 IfRRis>5&amp;<8

And EtC02 > 46 &amp; < 64

OrEtC02<24&amp;> 12

And SpO2 < 90% &amp; > 86% Then: 10 IfRRis>5&amp;<8

And EtC02 > 46 &amp; < 64

Or EtCO2 < 24 &amp; > 12

And SpO2 > 90% Then: 15 IfRR is >8 &amp;< 12And EtCO2>55Or EtCO2 < 18

And SpO2 < 90% Then: 20 IfRR is >8 &amp;< 12AndEtCO2>55OrEtCO2<18

And SpO2 > 90% Then:

Index = 2 down

Index - 5 down

Index = 6 down

Index = 6 down

Index = 7 down. 25 PI &lt;8 and no indication of hyperventilation or hypoventilation:

IfRRis>12 &amp; < 1630 AndEtCO2>64

OrEtCO2&lt;12

And SpO2 &lt; 90% Then:

IfRR is >12 &amp; < 1635 AndEtCO2>64

Or EtCO2 &lt; 12

And SpO2 > 90% Then:

If RRis>12 &amp; < 1640 And EtC02 > 44 &amp; < 64

OrEtCO2<24&amp;> 12

And SpO2 &lt; 90% Then:

IfRR is >12 &amp;< 1645 And EtCO2 > 44 &amp; < 64

Or EtCO2 <24 &amp;> 12

And SpO2 > 90% Then:

Index “ 6

Index = 7

Index-7

Index = 8 50 57 PI <8 and Indication of Hyperventilation 5 If RR is >40 And HR is >110 Then: Index = 4 Up 10 IfRR is >40 And EtCO2 > 50 And HR is >110 Then: Index = 3 Up 15 If RRis> 40 And EtCO2>50 And HR is >110 And SpO2<90% Then: Index = 2 Up 20 IfRR is >40 And EtCO2>44&amp;<50And HR is >100 And SpO2 < 90% Then: Index = 3 Up 25 IfRR is >40 And EtCO2 > 44 &amp; < 50And HR is >100 And SpO2>90% Then: Index = 4 Up If RR is > 32 &amp; <40 And HR is >100 Then: Index - 8 Up 30 IfRRis>32&amp;<40 And HR is >100 &amp; < 120And EtCO2>44&amp;<50 Then: Index = 7 Up 35 IfRRis>32&amp;<40 And HR is >100 And EtCO2>44 And SpO2 < 90% Then: Index = 6 Up 40 If RR is> 32 &amp; <40 And HR is >120 And EtCO2 > 55 Then: Index = 5 Up 45 IfRRis>28&amp;<32 And HR is >120AndEtCO2>55 Then: Index - 6 Up IfRR is> 28 &amp; <32 And HR is >100 And EtCO2 > 44 Then: Index = 7 Up 50 58

And EtC 02 > 44 &amp;< 64

OrEtCO2<24&amp;> 12

And SpO2 < 90% Then: Index = 7 up 5 IfRRis >16 &amp;<28And EtC02 > 44 &amp; < 64Or EtC02 <24 &amp; > 12

And SpO2 < 90% Then: Index = 7 up 10 15 Example 2 - Determining PI value using mathematical methods - Fuzzy logic algorithm A fuzzy logic inference is built using expert health care providers’ knowledgeand interpretation on anticipated PI values given four parameters: EtCC>2, Respiratory 20 Rate (RR), Heart Rate (HR) and SpO2.' Since there are differences in the normalranges of HR and RR between adults and children at different ages, the PI values forthe same parameters may differ. Hence ΡΙ-models are designed for distinct ages. Thefollowing descriptions apply to all modes, however the details are of adult mode. 25 In order to determine the PI value based on the four parameters (EtCO2, RR, HR and SpO2), a questionnaire is sent to 22 medical experts (nurses, respiratorytherapists, doctors and anesthesiologists). The questionnaire includes a set of 85 caseswith different parameters values. The experts are asked to assign a PI value accordingto a predefined code (Table 1). 30 PI Value Implication 10 Perfectly healthy normal 8 to 9 Normal 7 Understandable but requires more attention 5 to 6 It is recommended to pay more attention 3 to 5 ’Requires more attention and intervention is recommended Ito 3 Requires intervention

Fuzzy logic inference is built using the consensus of experts’ data. The designis carried out using matlab fuzzy logic toolbox (Mathworks Inc.). Initially, 35 membership functions are assigned for each parameter, determining the ranges for:Normal, High and Low values for HR and SpC>2 and Normal, High, Very-High, Lowand Very-Low for EtCC>2 and RR. The PI has ten membership functions, one for eachindex value. 40 A rule set is determined relating the inputs to the output, using the verbal descriptor of the membership functions (for example, if (ETCO2 is Very High) and(RR is Very High) and (SpC>2 is Normal) and (HR is High) then (PI is 2)). The rulesare summarized in a table 2. 59 RR EICO2 VH H N L VL SpO2 N, HR N SpO2 L 1=1-1 SpO2 VL -> 1=1 VH 3 6 4 3 2 HR L l=l-2 HR H -> |«H H 7 6 7 6 3 N 7 β 10 7 6 L 5 6 8 4 2 VL 3 3 4 2 2

The fuzzy logic operators used are: min for AND, max for OR, max forAggregation and centroide of area for de-fuzzification. 15

The examples described above are non-limiting examples and are not intendedto limit the scope of the disclosure. The described examples may comprise differentfeatures, not all of which are required in all embodiments of the disclosure. 20 60 CPEWan , crnxan ;w:n ατα imab^n pnow ρηνη irn ητ qaoa,Ρ’ηη ηχζοη paoana naa’na na^maa np’-ion .zrtwan rwaa mpAan pmib oxnnzj . Gtoca uer &amp;#9633;innn pv »*···»· · gm J«w\ _ _

♦ ,.~. Q9 «3/2)12 103236 4B ID .(n^oia nannn) cras&amp;amp;'an nwa

<img img-format="tif" img-content="drawing" file="IL205778AD00021.tif" id="idf0001" />