Patient data mining for cardiology screening

Description of equivalent WO 03040988 A2

PATIENT DATA MINING FOR CARDIOLOGY SCREENING

Cross Reference to Related Applications

This application claims the benefit of U.S. Provisional Application Serial No.

60/335,542, filed on November 2, 2001 , which is incorporated by reference herein in

its entirety.

Field of the Invention

The present invention relates to medical information processing systems, and,

more particularly to a computerized system and method for screening patients for

coronary heart disease (CHD), assessing a risk factor for a person to develop CHD and

managing a person with CHD.

Background of the Invention

Coronary heart disease is the number one killer in the western world. By

detecting coronary heart disease as early as possible, appropriate, effective, and cost-

effective treatment can be implemented.

However, cardiologists are faced with an ever-growing amount of data coming

from a variety of different sources: imaging modalities, patient reports, ECG tracings,

etc. As the number of information sources expand, extracting and assimilating all

available data manually, and assessing various treatment options, becomes more and

more difficult. Furthermore, with the push from managed care, cardiologists are

expected to treat and manage more patients in the same amount of time.

Currently, there is considerable evidence that cardiovascular risk and disease is

under-treated. Factors that account for this include gaps in knowledge, confusion over recommendations including target levels for lipids during treatment, poor

doctor-patient communication, and variations in physician understanding and

utilization of guidelines.

In view of the above, there exists a need for improved systems and methods

for screening persons for coronary heart disease, assessing the risks of individuals

patients in developing coronary heart disease, and managing patients with coronary

heart disease.

Summary of the Invention

A system and method for screening, detecting and managing patients with

coronary heart disease (CHD) is provided.

According to one aspect of the present invention, a method for screening for

coronary heart disease is provided including the steps of retrieving a test for

assessing risk of coronary heart disease, the test including a plurality of data fields

relating to coronary risk factors; accessing a database to populate the data fields of

the test with patient information of an individual patient, the database including

computerized patient records; and calculating a risk assessment of the individual

patient developing coronary heart disease. The method further includes the steps of

data mining information relating to the coronary risk factors from structured and

unstructured data sources; and compiling the information as a structured computerized pa

According to another aspect of the present invention, a coronary heart disease

screening system includes a first database including a plurality of structured

computerized patient records; a second database including a knowledge base relating

to coronary heart disease, the second database including at least one test for determining coronary heart disease risk wherein the at least one test includes a

plurality of data fields relating to coronary risk factors; and a processor for retrieving

the at least one test from the second database, populating the data fields of the at

least one test with patient information retrieved from the first database and

calculating a risk assessment for at least one patient. The first database is compiled by

data mining information relating to the coronary risk factors from structured and

unstructured data sources.

According to a further aspect of the present invention, a program storage

device readable by a machine, tangibly embodying a program of instructions

executable by the machine to perform method steps for screening for coronary heart

disease is provided. The method steps include retrieving a test for assessing risk of

coronary heart disease, the test including a plurality of data fields relating to coronary

risk factors; accessing a database to populate the data fields of the test with patient

information of an individual patient, the database including computerized patient

records; and calculating a risk assessment of the individual patient developing

coronary heart disease.

Brief Description of the Drawings

The above and other aspects, features and advantages of the present

invention will become more apparent from the following detailed description when

taken in conjunction with the accompanying drawings in which:

FIG. 1 is a block diagram of a computer processing system to which the

present invention may be applied according to an embodiment of the present

invention; FIG. 2 illustrates an exemplary coronary heart disease screening system

according to an embodiment of the present invention; and

FIG. 3 illustrates a flow diagram for screening, monitoring and managing a

patient according to an embodiment of the present invention.

Description of Preferred Embodiments

To facilitate a clear understanding of the present invention, illustrative

examples are provided herein which describe certain aspects of the invention.

However, it is to be appreciated that these illustrations are not meant to limit the

scope of the invention, and are provided herein to illustrate certain concepts

associated with the invention.

A system and method for screening, detecting and managing patients for

coronary heart disease (CHD) is provided. According to an embodiment of the present

invention, a computer-based coronary heart disease screening system will aid a

physician in the assessment and management of coronary heart disease. First, the

system will assimilate information from both imaging and non-imaging sources

within a computerized patient record (CPR). These data can be automatically

extracted, combined, an analyzed in a meaningful way, and the results presented to

the physician. Such a system will also help avoid mistakes, as well as provide a

novice with knowledge "captured" from expert users based on a domain knowledge

base of a disease of interest and established clinical guidelines. Within each specific

diagnostic test, the system will assist in automatically extracting information resulting

in potential improvements to workflow as well as providing a powerful "second

reader" in the evaluation of the results. Following evaluation, the system will also provide suggested therapies and follow-ups based on clinical guidelines. Finally, the

system could track the patient over time, assessing the progress of the disease and

the efficacy of therapy.

In the area of coronary artery disease, the world can be divided into two

groups: those with known or suspected coronary artery disease, and those without.

In the latter case, the key is to promote prevention and decrease the risk of coronary

artery events. Here, the coronary heart disease screening system is targeted to the

clinical cardiologist, and the general practitioner, to help assess, monitor, and reduce

the risk of coronary heart disease.

In the case of people with known or suspected coronary heart disease, the role

of a computer-aided coronary heart disease screening system is slightly different.

First, such a system could aid in the assessment and diagnosis of the disease by the

physician. Next, the system could help a cardiologist assess the severity of the

disease, and help identify potential therapies. Finally, the system could assist with

assessing the progression or regression of the disease either over time or in response

to therapy.

It is to be understood that the present invention may be implemented in

various forms of hardware, software, firmware, special purpose processors, or a

combination thereof. Preferably, the present invention is implemented in software as

a program tangibly embodied on a program storage device. The program may be

uploaded to, and executed by, a machine comprising any suitable architecture.

Preferably, the machine is implemented on a computer platform having hardware

such as one or more central processing units (CPU), a random access memory (RAM),

and input/output (I/O) interface(s). The computer platform also includes an operating system and microinstruction code. The various processes and functions described

herein may either be part of the microinstruction code or part of the program (or

combination thereof) which is executed via the operating system. In addition,

various other peripheral devices may be connected to the computer platform such as

an additional data storage device and a printing device.

It is to be understood that, because some of the constituent system

components and method steps depicted in the accompanying figures are preferably

implemented in software, the actual connections between the system components

(or the process steps) may differ depending upon the manner in which the present

invention is programmed.

FIG. 1 is a block diagram of a computer processing system 100 to which the

present invention may be applied according to an embodiment of the present

invention. The system 100 includes at least one processor (hereinafter processor)

102 operatively coupled to other components via a system bus 104. A read-only

memory (ROM) 106, a random access memory (RAM) 108, an I/O interface 1 10, a

network interface 1 12, and external storage 1 14 are operatively coupled to the

system bus 104. Various peripheral devices such as, for example, a display device, a

disk storage device(e.g., a magnetic or optical disk storage device), a keyboard, and a

mouse, may be operatively coupled to the system bus 104 by the I/O interface 110 or

the network interface 112.

The computer system 100 may be a standalone system or be linked to a

network via the network interface 112. The network interface 112 may be a hard¬

wired interface. However, in various exemplary embodiments, the network interface

112 can include any device suitable to transmit information to and from another device, such as a universal asynchronous receiver/transmitter (UART), a parallel digital

interface, a software interface or any combination of known or later developed

software and hardware. The network interface may be linked to various types of

networks, including a local area network (LAN), a wide area network (WAN), an

intranet, a virtual private network (VPN), and the Internet.

The external storage 1 14 may be implemented using a database management

system (DBMS) managed by the processor 102 and residing on a memory such as a

hard disk. However, it should be appreciated that the external storage 1 14 may be

implemented on one or more additional computer systems. For example, the

external storage 1 14 may include a data warehouse system residing on a separate

computer system.

Those skilled in the art will appreciate that other alternative computing

environments may be used without departing from the spirit and scope of the

present invention.

Referring to FIG. 2, an exemplary coronary heart disease (CHD) screening

system 200 according to an embodiment of the present invention is illustrated. The

CHD screening system 200 includes a processor 202 which processes a plurality of

modules for performing different tasks. The processor is coupled to a first database

204 compiled to store a plurality of structured computerized patient records (CPR)

relating to a disease of interest, here, coronary heart disease.

Preferably, the structured database 204 is populated with population-based

patient information using data mining techniques described in "Patient Data Mining,"

by Rao et al., copending U.S. Patent Application Serial No. 10/ , , (Attorney

Docket No. 8706-600) filed herewith, which is incorporated by reference herein in its entirety. That patent application teaches a data mining framework for mining high-

quality structured clinical information. The data mining framework includes a data

miner 206, having functions and capabilities as in the REMIND system, commercially

available from Siemens Medical Solutions, that mines medical information from

computerized patient records (CPRs) based on domain-specific knowledge contained

in a knowledge base. The CPRs may be of structured 208 (e.g., chart, tables, billing

information, etc.) and/or unstructured formats 210 (e.g., doctors' dictations, images

such as MR (magnetic resonance) images and CT (computerized tomography)scans,

ECG waveforms, etc.). The domain-specific knowledge may relate to a disease of

interest, a hospital, etc. The data miner 206 includes components for extracting

information from the CPRs, combining all available evidence in a principled fashion

over time, and drawing inferences from this combination process. The mined medical

information is stored in the structured CPR database, such as database 204.

The processor 202 is further coupled to a second database 212 including state

of the art information relating to the disease of interest. This information may include

standard procedures, established guidelines for treatment, standardized tests for

assessment and diagnosis, etc.

Additionally, the processor 202 is adapted to receive manually inputted patient

data 214 which it will process and store in the first structured database 204.

The CHD screening system 200 interacts with the first structured database 204

and the medical knowledge database 212 to assess the risk of a patient developing

CHD, to recommend therapies and lifestyle changes to reduce the patient's assessed

risk, and to perform sensitivity analysis to determine what factors are of the greatest

risk to a patient. Each task performed by the CHD screening system 200 is performed by an executable module residing either in the processor of the system 202 and/or in

a memory device (e.g., RAM, ROM, external storage, etc.) of the system.

Referring to FIGS 2 and 3, the CHD screening system will be further described

along with methods for assessing CHD risk, monitoring CHD patients and suggesting

therapies and lifestyle changes.

The goal of both primary and secondary CHD prevention is to decrease the risk

of subsequent acute coronary events, and thereby decrease mortality and prolong

survival. In primary prevention, the idea is to identify healthy (asymptomatic)

individuals at high risk of developing coronary artery disease, and initiate therapies

and lifestyle changes to lower this risk. Secondary prevention does the same for

people who have had an acute coronary event. The system and method of the

present invention will assist physicians and play a significant role in assessment of

risk, targeted suggestions for therapy and lifestyle changes based on established

guidelines, and monitoring patient progress towards goals of risk reduction.

In primary prevention, asymptomatic individuals at high risk of developing

coronary artery disease are identified, and therapies initiated and lifestyle changes

recommended to lower this risk. First, asymptomatic patients are assessed for risk for

coronary heart disease on the basis of risk factors. For example, the National

Cholesterol Education Program (NCEP) has recently produced a set of Adult

Treatment Panel (ATP III) guidelines for the treatment and management of lipid

disorders. In these guidelines, the risk of an acute coronary event over a 10-year

period for people with no history of coronary heart disease is calculated based on a

modified version of the Framingham algorithm or test. The Framingham algorithm

uses traditional risk factors, such as gender, obesity, smoking, total cholesterol, HDL- cholesterol, age, diabetes, and blood pressure, to determine overall risk. The ATP III

guidelines also include other risk factors, such as family history and hypertension, in

its risk model. In addition to these traditional risk factors, the guideline

acknowledges emerging risk factors, such as hemocysteine, lipoprotein(a), and

inflammatory markers such as high sensitivity C-reactive proteins which can indicate

risk of acute coronary events. In addition, the NCEP ATP III has developed specific

recommendations for therapy and lifestyle changes based on these risk factors for

both primary and secondary prevention. Such tests, recommendations and guidelines

will be stored in the medical knowledge database 212.

These guidelines serve to help the clinical cardiologist, as well as the patient's

primary-care physician, assess the risk and help prevent the incidence of an acute

coronary artery event. In this situation, the CHD screening system 200 could provide

valuable assistance in a variety of different ways. First, the assessment of risk

requires obtaining clinical information contained in a variety of different locations

within a patient's record. Structured 208 and unstructured 210 data are mined via

the data miner 206 and stored in structured CPR in database 204 (step 302). The

system 200 then accessed the second database 212 to retrieve a test, such as the

Framingham algorithm, to assess the risk of an individual patient (step 304). An

assessment module 202-1 receives the test and populates a plurality of data fields

within the test with information retrieved from the structured database 204 and

calculates the risk for the patient (step 306). Depending on the risk assessment or

score, the patient is categorized as being a low risk (step 308), intermediate risk

(310) or high risk (step 312). It is to be appreciated that all the information necessary to calculate the risk

assessment may not be readily available from the patient record. If the information is

missing, the system will mine available data to make a probabilistic assertion

regarding the missing information. The system will then calculate the patient's risk

assessment with the probabilistic information.

It is to be appreciated that the ability to automatically retrieve information and

calculate these risks would save the physician time, and would enable automated

screening of a very large population. Incomplete or conflicting information could be

brought to the attention of the physician.

Once the risk is assessed for an individual patient, a specific set of

recommended tests, therapies and/or lifestyle changes, targeted to the patient and

based on specific adopted guidelines, will be automatically generated (step 316).

Upon subsequent screening visits by the patient, the CHD system 200 will track

recommendations against patient performance via a patient management module

202-2. For example, the patient's cholesterol levels and lifestyle changes in

subsequent check-ups could be matched against guideline targets and previously

recorded levels, and follow-up reports automatically generated for the physician. The

patient management module may periodically reassess the risk of the patient and

alert the appropriate personnel if a significant change is detected. Additionally, the

patient management module may recommend a time for a reassessment due to the

length of time between assessments and or tests.

In addition to assessing current risk, the CHD system 200 will perform

sensitivity analysis on the various risk factors to assess the importance of each

individual risk factor on that patient via a sensitivity module 202-3. That is, for each risk factor, the system will weigh its individual importance in assigning the overall risk

of acute coronary events to the individual. First, a model is created to simulate a

patient with similar characteristics of the patient being tested (step 320). A modeling

module 202-4 generates the model either by simulating a similar patient based on

the data stored in the medical knowledge database 212 or by mining data of similar

patients from population-based data sources via the data miner 206 using a domain

knowledge base of the disease of interest (i.e., coronary heart disease; alternatively,

the model could be a combination of both. The sensitivity module 202-3 then

interacts with the modeling module 202-4 by varying input data to simulate different

scenarios to determine which factor most influences the risk assessment of the

patient (step 322).

The sensitivity analysis is important for several reasons. First, for many

patients, not all of the risk factors are usually tested. For example, one of the risk

factors is the presence of diabetes. However, not all patients have had a blood-sugar

test done. In this case, the risk of heart disease is first assessed without this

information. Then, an analysis is done with different blood-sugar results to see

whether the risk assessment outcome changes. If the change is significant, the

system may recommend that a blood-sugar test be done to refine the cardiac risk for

the patient. Secondly, the sensitivity analysis will assess when the patient needs to

come back for another screening, and what exams should be performed (step 316).

For example, it is known that blood pressure can change significantly from reading to

reading, and it can also slowly go up or down over time. By knowing the typical

variability of such a variable, the system can help decide when the patient needs to come back for another blood pressure reading by assessing what type of change

would result in a significant change in risk assessment.

Additionally, the system 200 will recommend whether further risk

stratification is needed. If the initial risk assessment shows that a person has a low

risk of acute coronary events (step 308), then a physician may recommend a healthy

lifestyle with diet and exercise (step 316). Conversely, if the assessment shows a

high risk of acute coronary event (step 312), then a physician may decide on some

kind of therapy (316), such as aspirin or cholesterol-reducing drugs, based on clinical

guidelines. However, some people will show an intermediate risk of heart disease

(step 310). In these cases, there may be a need to further assess and refine.the risk

of coronary heart disease in a patient (step 314).

Where further risk stratification is needed, a number of different techniques

can be used, and the choice of a technique may depend on the cardiologist's

experience, comfort level, and access to equipment. Many of the techniques

developed to further stratify risk do so by measuring artherosclerotic burden, for

example, (1 ) measurement of ankle/brachial blood pressure index (ABI); (2)

measurement of hemocysteine, lipoprotein(a), and inflammatory markers such as

high sensitivity C-reactive proteins, as well as other emerging biochemical markers;

(3) measurement of intima-media thickness (IMT) from the carotid arteries using

high-frequency B-mode ultrasound; (4) assessment of plaques in coronary arteries

using Electron-beam Computed Tomography (EBCT); (5) assessment of composition

of artherosclerotic plaque with magnetic resonance imaging (MRI); (6) assessment of

endothelial function to determine artherosclerotic risk; and (7) scoring coronary

calcium, e.g., using the Agatston score. To facilitate the risk stratification, the system 200 may further include an

imaging module 202-5 to automatically extract information from the imaging sources

mentioned above (e.g., by conventional image segmentation methods), and combine

the extracted information with the previously assessed risk to reassess the overall risk

of the patient. The results of any risk stratification could be used to generate patient-

directed recommenda'tions based on established clinical guidelines using this

additional risk assessment (step 316).

Implementation of the CHD system for secondary prevention is a simpler than

for primary prevention. The reason is that once a patient has had an acute coronary

event, they will always be at high-risk for a subsequent event. Therefore, there is

little need for risk assessment in these individuals. Rather, the main emphasis in

secondary prevention is to create a specific set of recommended therapies and

lifestyle changes, targeted to the patient and based on specific adopted guidelines.

Goals for lifestyle changes as well as lipid management and blood pressure

have been established for primary and secondary prevention. By tracking a patient

over time (step 318), the system could automatically assess whether the patient is

achieving the desired goals for risk reduction, and whether changes need to be

implemented either in the therapy or implementation. The effects of specific diet

changes, exercise, or cholesterol-lowering drugs, for example, can be feedback into

the system to redesign therapies and create new recommendations for individual

patients. Conventionally, tracking patients require manual monitoring of patient

information, and comparing against established standards. These manual monitoring

techniques have resulted in inconsistent management of cardiovascular risk. Furthermore, the CHD system can be used to assist in diagnosis of a patient

with CHD. Often times, the first time a patient is referred to a cardiologist is after

coronary artery disease has significantly progressed, and the patient exhibits some

symptoms. For this group of people, it is important to be able to diagnose the

disease, and then apply appropriate therapy and monitor their progress in a rapid

manner. The diagnosis may be performed combining all available information about

the patient and perform a probabilistic inference on patient-specific issues based on

the domain knowledge base using techniques described in "Patient Data Mining for

Diagnosis and Projections of Patient States," by Rao et al., copending U.S. Patent

Application Serial No. 10/ , , (Attorney Docket No. 8706-624) filed herewith,

which isjncorporated by reference herein in its entirety.

For example, in an emergency room, patients may present with chest pain.

The emergency room physician must be able to diagnose acute coronary events, and

may need to initiate therapies to stabilize the patient. According to ACC/AHA

guidelines, electrocardiography (ECG) is the procedure of first choice in patients

presenting with chest pain, dizziness or syncope - symptoms that may be predictive

of sudden death or myocardial infarction. In situations where the ECG is non-

diagnostic, ultrasound can be used to assess regional systolic wall motion

abnormalities. Since the emergency room physician may not be as experienced as a

cardiologist to interpret these tests, the CHD system can provide a checklist of items

to assist with diagnosis, and then automatically extract information from sources,

such as the ECG or ultrasound exams, to assist in rapid determination of an acute

coronary event. In addition, the system could provide suggested immediate

therapies based on established clinical guidelines. Furthermore, the CHD system could aid a clinical cardiologist in answering

important clinical questions, including: diagnosis of obstructive coronary heart

disease; assessment of severity of disease and complications; assessment of viability

of diseased heart tissue; and recommendations for patient management based on

established clinical guidelines.

A number of diagnostic tools are at the cardiologist's disposal to help answer

these questions, e.g., electrocardiography, coronary angiography, radionuclide

imaging, ultrasound, magnetic resonance imaging, electron-beam computed

tomography, etc. Each of these modalities measures either direct or surrogate

indicators of coronary artery disease. Individually, each can help provide evidence of

coronary artery disease. The choice of diagnostic tool used by the cardiologist is often

made based on availability, experience, and comfort level. Each modality measures

something slightly different in assessing coronary artery disease. Potentially more

powerful, therefore, is the registration of data from different sources to provide a

more complete picture in assessing coronary artery disease. Currently, diagnosis of

coronary artery disease is often done using a qualitative, or semi-quantitative,

approach. As a result, the effectiveness of such diagnostic approaches depends to a

great extent on the experience and knowledge of the doctor. For example, stress

echocardiography for. assessment for global function and regional abnormalities is

done using a visual inspection followed by point scoring.

The CHD system will extract and combine information in a quantitative

manner from a variety of different sources to help the clinical cardiologist address

these clinical questions, augmenting the physician's own intuition and experience. In this manner, the system would assist the physician in their own decision-making

process, following accepted guidelines and practices.

In addition to detecting coronary artery disease, a number of imaging

modalities can be used to assess the progression or regression of the disease either

over time or in response to therapy. Some of these include ultrasound, coronary

angiography, radionuclide imaging, and intravascular ultrasound. Many times, these

techniques are used to study the effects of specific therapy, such as revascularization.

In another scenario, these techniques could be used to monitor the progression or

regression of a patient over time to assess when and if intervention is necessary.

The CHD system will extract information from the images produced, e.g., by

segmentation, volume rendering, etc., and register the information on a patient from

different points of time and from different sources, to assess the progression or

regression of disease. By creating such an automatic system, physicians can more

easily monitor the progression or regression of coronary artery disease, which can

assist in deciding the efficacy of a particular plan of treatment.

In the area of coronary artery disease, the systems and methods of the present

invention can potentially play a large role in the total management of a patient,

including prevention, detection, therapy, and monitoring. Today, information about

the patient comes a wide variety of different sources, including patient clinical

history, waveform data such as ECG, imaging data, blood tests, etc. Furthermore,

numerous clinical guidelines are established by bodies such as the ACC, AHA, and ESC

to discuss issues such as prevention, detection, and therapy. The system and method

of the present invention can assist physicians by automatically collecting information

from a wide variety of different sources and analyzing them. Information can be presented to the physician along with suggestions based on established clinical

guidelines.

It is to be appreciated that various embodiments of the present invention are

to be defined in the context of the physician's workflow. Such embodiments could

exist as a distributed system within different sub-systems as defined by clinical

workflow and usefulness. For example, some components may fit within the

imaging modality, such as on the ultrasound system or on an MRI console system.

Other pieces or components may reside on a review workstation, like a KinetDx^® or

Leonardo™ workstation. A comprehensive system may belong on a Sorian™

cardiology system. Together, they will form a united clinical solution. Alternatively,

such a system could exist as a remote server resulting in an ASP(Application Service

Provider)-model solution. This could allow small systems, such as hand-held

ultrasound systems and other hand-held devices (e.g., personal digital assistants,

handheld computers, laptop computers, etc.) to leverage the CHD system at a remote

site, in an emergency room or at the scene of an incident outside the hospital.

Although illustrative embodiments of the present invention have been

described herein with reference to the accompanying drawings, it is to be understood

that the invention is not limited to those precise embodiments, and that various

other changes and modifications may be affected therein by one skilled in the art

without departing from the scope or spirit of the invention.