Clinical tool for structure localization

The present invention relates to methods of inducing medical intervention using anatomical or pathological landmarks, and more particularly to data collection and processing methods that facilitate image-guided surgery.

Minimal invasive techniques have been developed for various surgical procedures. During these procedures, clinicians typically rely on intraoperative imaging to direct the device to a targeted treatment or diagnostic site. This operation is often the most time-consuming part of the procedure, especially when the intraoperative image is a two-dimensional image, as it can require considerable skill and intuition. With the increasing use of minimally invasive techniques, clinical tools are needed to help clinicians locate the target site quickly and accurately during surgical procedures.

<p>The present invention provides a system that assists a clinician in localizing a structure in a patient's body during a medical intervention.</p>

<p> The method generally involves first generating a set of measurements that define the location of one target site relative to at least two reference structures or at least one reference structure and local coordinate system. The set of measurements is then used during the surgical procedure to project the estimated location of the target site onto the intraoperative image.</p><p> The method provided by the present invention uses a data collection method for acquiring triangular parameter data performed prior to a surgical procedure and previously acquired triangulation parameter data on an intraoperative image. Includes image-guided localization methods for plotting estimated target structure locations. The data collection method serves as a reference structure for use in localizing the target structure, forming a spatially accurate 3D image volume of the body part, where one target structure for the surgical procedure is located inside. Select the structure (s) in the image volume to fulfill the above, find the landmark point coordinates that define the respective locations of the target structure and the reference structure in the volume, and determine the location of the landmark point. Includes calculating associated triangle parameter data.</p><p> The localization method performed during the surgical procedure is to position the reference structure on the medical image obtained during the operation and to use the calculated triangular parameter data and the identified reference structure location to perform the intraoperative image. Includes plotting the estimated location of the target structure above.</p><p> In an embodiment that utilizes 3D imaging during a surgical procedure, previously determined triangular parameters based on 3D landmark point coordinates are used to estimate the target structure on a 3D intraoperative image in which the location of the reference structure is located. Used to plot the location to be.</p><p> In embodiments that utilize 2D imaging during a surgical procedure, the method of generating triangular parameter data is to calculate the isocenter of the body part and to accommodate landmark points around the isocenter axis, corresponding to the selected 2D plane view. Further includes rotating to a given angle, projecting a landmark point on a selected plane, and calculating the triangular parameters that relate the location of the projected landmark point coordinates. During the surgical procedure, the estimated location of the target structure is plotted on selected 2D plane images acquired intraoperatively. The plotted locations are based on the projected triangular parameter data and the identified locations of the reference structure in the 2D intraoperative image.</p><p> In some embodiments, a single reference structure and local coordinate system are used with one target structure to determine the set of localization parameters in the 2D imaging plane. The target structure is plotted on the intraoperative image based on the identified location of the reference structure and the localization parameters that associate the location of the reference structure, the target structure, and the local coordinate system axes.</p>

The present invention provides clinicians with a useful tool for localizing internal structures, especially during minimally invasive procedures where the clinician does not have a direct line-of-sight view of one target structure. The target structure may be an anatomical structure, a pathological feature, or an implantable device. During a minimally invasive procedure, the clinician usually guides the device or medical device towards the target structure by observing the position of the device in a 2D or 3D image obtained intraoperatively. These procedures are time consuming and can require considerable skill and intuition. The present invention provides a method of plotting the estimated location of a target structure on an intraoperative medical image to assist the clinician in localizing the target structure. The plotted locations are estimated based on the triangular parameter data collected prior to associating the target structure location with respect to the selected reference structure. Triangular data can be obtained from a 3D image of a patient undergoing procedure or from another subject or population of subjects.

The methods provided by the present invention include predictive data acquisition methods and localization methods for use during surgical procedures. A data collection method is performed to determine the triangular parameters that associate the location of one structure targeted by the surgical procedure with respect to the location of multiple selected reference structures. Triangular parameters are later applied during surgical procedures in a localization method for guiding the instrument to the target structure. The localization method may be performed using either 3D or 2D intraoperative imaging. Data acquisition methods are performed using 3D imaging, and the steps performed will depend on whether 3D intraoperative images are used during the surgical procedure or 2D intraoperative images are used. In the following description, data acquisition methods and localization methods for 3D image-guided localization applications will be described in relation to FIGS. 1 to 3B. Data acquisition and localization methods for 2D image-guided localization applications are described in relation to FIGS. 4-8.

FIG. 1 is a flow chart summarizing the steps involved in the data collection method for calculating the triangular parameter data that associates the 3D reference structure location with the target structure location. Triangular parameter data collected using method 10 in FIG. 1 will be used later in an image-guided procedure using 3D intraoperative medical images. Based on the triangular parameter data collected, one target site location will be plotted within the 3D volume defined by the 3D medical image acquired during the surgical procedure. Intraoperative images are usually obtained by fluoroscopy, but can also be obtained using any available medical imaging technique, such as ultrasonography.

In step 15 of data acquisition method 10, using any existing type of imaging method, examples include, but are not limited to, computed tomography (CT) and magnetic resonance imaging (MRI). , Volume medical images from patients or applicants are collected. In some applications, the image is a time-gate controlled image to reduce physiological motor artifacts from the triangular measurements. Images may be gated at specific points in the respiratory cycle and / or cardiac cycle to reduce the effects of respiratory and cardiac movements on triangular parameter measurements.

Once a series of volume images is acquired, the images are stacked to form a spatially accurate 3D image volume in step 20. A medical image analysis software package called Analyze, developed at Mayo Clinic in Rochester, Minnesota and sold by AnalyzeDirect, may be used to carry out Step 20. The image volume acquired in step 20 serves as a virtual representation of the true anatomy found in the body of the patient or volunteer from whom the image is taken. The image volume is taken to include the target structure and the selected reference structure. It is expected that at least three reference structures will be required to generate a set of triangular parameter data that can be used to plot the location of one target structure on a 3D intraoperative image.

The reference structure may be selected as a structure that is generally more easily positioned and identified on the medical image as compared to the target structure. The smaller the anatomical location variation of the selected reference structure with respect to each other and with respect to the target structure, the more accurate it is when plotting the estimated location of the target structure during the surgical procedure. There will be. In this data acquisition phase, multiple reference points are initially selected and some may be later eliminated due to the large anatomical variation that causes undesired volatility of the triangular parameters.

The reference structure is not limited to the anatomical structure. An embedded device may be selected as the reference structure. Implantable devices, such as pacemakers, drug pumps, leads, orthopedic inserts, or other devices, may already be present in the patient's body. The implantable device may be installed during the image guidance procedure to serve as a reference structure. For example, a catheter tip or other type of marker may be placed at the desired location to serve as a reference structure.

As mentioned earlier, the target structure may be an anatomical structure, a pathological feature, or an implantable device. Medical interventions remove, reposition, or relocate existing implantable medical devices to remove tissue, remove tissue, deliver drugs or biological substances, implant new medical devices, or It may be performed to modify or adjust in other ways, or to perform other procedures at the target site.

Steps 25-35 are intended to define landmark point coordinates that define the respective locations of the reference structure and the target structure. Many methods may be used to define landmark points for each structure. In one embodiment, the user may use an instruction tool to simply select a point in the medical image that corresponds to the structure. In that case, the coordinates of the selected point will be defined as the landmark point coordinates for the structure. The points selected may be identifiable aspects of the structure and can be selected in a reproducible manner to improve the accuracy of the plotted target structure locations during the surgical procedure. Landmark points may otherwise be selected using automated techniques that identify a structural point in a digitized grayscale or color image. In another embodiment, as shown in FIG. 1, the landmark point coordinates may be determined by first generating a slope containing the structure where the landmark points are determined, passing through the 3D image volume. Good. "Oblique" found in the Analyze software packageThe "Sections)" module may be used to perform this step. A plane is created through the target structure. Additional planes are generated that pass through each of the selected reference structures. In one specific application, triangular parameter data collection for procedures requiring cannulation of the coronary sinus ostium (CSos) produces one plane through the target CSos, and, for example, the superior vena cava. It may include producing two planes through two reference structures that can be selected as the (SVC) mouth and tricuspid valve (TV) ring.

Slopes are created through 3D volumes so that each plane best represents the 2D shape of its target and reference structure. In the example shown above, the plane of the SVC mouth will be generated around the SVC, including where the SVC enters the right atrium. The plane of the TV ring is generated to include the perimeter of the valve at the interface between the right atrium and the right ventricle, and the plane of the CSos is generated to include the perimeter of the CSos along the wall of the right atrium. Will.

For the target structure and each reference structure, once the planes are generated, the points that define the relevant structures are selected in step 30. In one embodiment, points along the boundaries of each structure may be manually selected using an instruction tool. In other embodiments, point selection may be automated. For example, points may be automatically selected based on the boundary detection of a digitized grayscale or color image.

The point picking tool in the "Slanted Section" module of the Analyze software may be used to perform step 30 and each set of selected boundary points or edge coordinates can be saved to a text file.

In step 35, landmark point coordinates are obtained for each reference structure and target structure. In the embodiment shown in FIG. 1, the landmark point coordinates for a given structure are obtained from one or more of the points selected in step 30. The landmark point coordinates define the location of the relevant structure within the imaging volume. In one embodiment, the landmark point may be the center of gravity calculated from points selected along the edges or perimeter of the structure. Landmark coordinates may be calculated using commercially available software such as Matlab software (Mathworks, Natick, Mass.). A simple Matlab script developed by the inventor, called "centroid.m", gets a scaled set of selected boundary point coordinates and the average location of these points (also known as the centroid). To calculate. Triangular parameters are obtained in step 37 for the landmark point coordinate set obtained in step 35, which corresponds to the target structure landmark point and the selected reference structure landmark point. The distance between the reference structure landmark points and the distance between the reference structure landmark points and the target structure landmark points are obtained. The distance ratio that correlates the distance between the reference structure and the target structure with respect to the distance between the two reference structures is then determined. By using the ratio of distance measurements, the spatial relationship of the structure can be estimated regardless of the unit of measurement or body size.

At step 37, the angle of the line interval extending between the landmark points is also determined. Using the calculated distance ratios and angles, the location of the target structure can be estimated and plotted in an image where the location of the reference structure is known.

The same Matlab script developed by the present inventor used to calculate the centroid can be used to calculate the distance between each of the landmark points. These distances define the length of each hypotenuse of the triangle defined by the landmark points of the target structure and the two reference structures. The calculated distance ratio and the angle between each of the three vector pairs formed by the landmark triangles can be determined. In one embodiment, the triangle parameter determined in step 37 includes at least two distance ratios and two angles determined from two different triangles, each containing two reference landmark points and a target landmark point.

Method 10 can be performed in a patient-specific manner, and method 10 is performed in anticipation of the patient undergoing the procedure. The previously determined triangular parameters for the patient are applied during the procedure to plot the estimated location of the target structure on the intraoperative image taken, where the reference structure location has been identified.

Method 10 can be performed on a large number of subjects selected from the population so that representative triangular parameters for a set of reference and target structures can be determined. In a given patient undergoing an image-guided procedure, less volatile triangular parameters can be used in the patient population to help localize the target structure. Triangular parameters obtained from a large number of subjects may be statistically analyzed to obtain a representative set of triangular parameters. For example, the average value of the triangular parameters may be obtained. Triangular parameter data obtained from the patient population is readily available for use during emergency procedures where predictive data acquisition imaging procedures are not allowed to be performed in a patient-specific manner in time. be able to.

Figures 2A and 2B show a set of triangular parameters that can be calculated during the data collection method 10 (Figure 1). These parameters will be used in localization methods during image-guided surgical procedures that utilize 3D intraoperative images. It is expected that at least three reference points will generally be needed to determine the triangular parameters that allow one target structure location to be plotted in the 3D intraoperative image. In Figure 2A, three reference landmark points 71, 72, and 73 and target structure landmark points 75 are identified. The two reference landmark points 71 and 72 and the target landmark points 75 form the triangle ABC. The set of triangle parameters obtained from the triangle ABC will include one distance ratio and one angle. The distance ratio will be the ratio of the distance between either the reference landmark points 71 or 72 and the target landmark point 75 to the distance between the two reference landmark points 71 and 72. For example, the distances AC and AB may be obtained by calculating the distance ratio AC: AB from the landmark point coordinates. The angle (angle 1) surrounded by the sides AC and AB of the triangle is also calculated from the landmark point coordinates.

In FIG. 2B, a second set of triangular parameters is obtained from the triangular DBC formed by the third reference landmark 73, the landmark 72, and the target landmark 75. The second set of triangular parameters also includes the distance ratio and angle obtained from the triangular DBC. For example, the distance ratio DC: DB and angle 2 may be calculated from the landmark point coordinates.

Further reference structures may be selected to improve the accuracy of the one target point location to be plotted. In some 3D applications, triangle parameters may be collected for three or more triangles formed when four or more reference structures are selected. If three landmark points are identified for the three reference structures, then for the two triangles formed by the three reference landmark points and the target landmark points, a triangle containing one distance ratio and one angle, respectively. Two sets of parameters are required. In other embodiments, additional triangular parameters may be derived from the triangle formed using the four or more reference structure landmark points and the target structure landmark points. By increasing the number of reference structures used to obtain a set of triangular data, the error in predicting a single target site location can be minimized. Target site locations can be more accurately predicted and the overall procedure time can be predicted, although more time may be required during the invasive procedure to identify additional reference structures within the intraoperative image. decrease.

Other techniques for positioning reference or target structures may be relied upon during the 3D imaging procedure. For example, electrophysiological measurements may be performed to assist in positioning the electrophysiological structure within the heart. In another example application of the invention, the localization method described herein may be used to guide the clinician to place a pacing electrode on the bundle of His. Data acquisition procedures that rely on 3D imaging and electrophysiological mapping may be used to obtain the triangular parameters needed to associate the His bundle location with the selected reference structure. Electrophysiological measurements will be used to confirm the location of the His bundle.

FIG. 3A is a diagram of the steps involved in the method of applying 3D triangular parameters when plotting estimated target structure locations within a 3D intraoperative image. In some applications, three reference points that are coplanar with the target structure can be used to localize one target structure. The localization method shown in FIG. 3A shows one method of plotting the estimated target structure location in a 3D image based on the triangular parameters derived from the three coplanar reference points.

At step 102, the locations of the three reference structures 71, 72, and 73 are marked by the user within the intraoperative image frame 101. Image frame 101 is a 3D image of a volume that is expected to contain each of a reference structure and a target structure. The user can specify the location of the reference structure using an instruction tool, a touch screen, or other user equipment compatible with the imaging system. The user should attempt to mark the location of each reference structure at the point corresponding to each landmark point used when retrieving the triangular data, in order to improve the accuracy of the plotted target structure locations. For example, if the landmark point coordinates correspond to the centroid of the structure, the clinician should attempt to mark the centroid in image 101.

After marking the reference structures 71, 72, and 73, the distance between the reference structures 71 and 72 in the intraoperative image 101 is measured in step 104. Using the labeling arrangements shown in FIGS. 2A and 2B, the measured distance corresponds to the distance AB of one hypotenuse of the triangle ABC formed by the two reference structures 71 and 72 and the target structure. Distance AB can be measured in any unit, as a unitless distance ratio will be applied to estimate the location of the target structure.

One of several possible ways to visualize the mathematics behind the application of 3D triangle parameters is to think of the circle in the space defined by the identified reference point location and the triangle parameters. In step 106, the triangle parameter is applied to generate the circle 110. An orbit rotates around the line segment AB at an angle of 1, and a circle 110 whose circumference is located at a distance AC from the point 71 is defined. The distance AC is obtained from the previously obtained distance ratio AC: AB and the distance AB measured in image 101. Angle 1 is the previously determined angle as described in FIG. 2A.

The target landmark point is located on or approximately on the circle 110. The second set of triangular parameter data obtained during the data collection method 10 is used to estimate where the target landmark point is located on the circle 110. In step 108, the second circle 114 is generated by rotating the trajectory extending from the reference point 73 at an angle 2 from the line segment DB. The circumference of circle 114 is located at distance DC from point 73, as determined from the previously determined distance ratio DC: DB and the distance DB measured in image 101.

Target landmark points 76 are plotted at the intersection of circles 110 and 114. The plotted points 76 are displayed on the intraoperative image and provide a target for the clinician to guide the instrument during the invasive procedure. Circles 110 and 114 may or may not be displayed. The circle shown in FIG. 3A is shown to illustrate the calculations that can be used to identify the estimated target structure point 76.

The area or perimeter surrounding the plotted points 76 may be shaded or contoured on the image display to indicate the inherent error or volatility of the triangular parameters. Circles 110 and 114 may be defined by the thickness of the line producing the estimated 3D target volume at the intersection 76. The error may occur when the user-specified reference structure locations 71, 72, and 73 in image frame 101 do not exactly match the landmark point coordinates used when determining the triangular parameters. When the triangular parameters are obtained from the patient population, there will be constant inter-individual variability in the measured distance ratios and angles, resulting in variability in the expected target structure location.

In applications where the three reference points and target structures are not coplanar, the circles 110 and 114 generated using the two sets of triangular parameters may intersect at two or more points. At times, the clinician may be able to logically eliminate one intersection as not being located in an anatomically realistic location that corresponds to the target structure. If three points are not sufficient to localize the target structure, then a fourth reference structure is needed to generate yet another circle to determine the unique intersection that defines the target structure location. There is. The estimated target structure location will be plotted at the intersection of the three circles generated using the three sets of triangular parameters obtained using the four reference landmark points.

Figure 3B shows the use of four reference landmark points to plot estimated target structure locations within a 3D intraoperative image. At step 130, the locations of the four reference structures 71, 72, 73, and 74 are marked by the user within the intraoperative image frame 101. Image frame 101 is a 3D image of a volume that is expected to contain each of a reference structure and a target structure.

In step 132, the triangle parameter is applied to generate the circle 110. An orbit rotates around the line segment AB at an angle of 1, and a circle 110 whose circumference is located at a distance AC from the point 71 is defined. The distance AC is obtained from the previously obtained distance ratio AC: AB and the distance AB measured in image 101. Angle 1 is the previously determined angle as described in FIG. 2A.

The target landmark point is located on or approximately on the circle 110. The second set of triangular parameter data obtained during the data collection method 10 is used to estimate where the target landmark point is located on the circle 110. In step 134, a second circle 114 is generated by rotating the trajectory extending from the reference point 73 at an angle 2 from the line segment DB. The circumference of circle 114 is located at distance DC from point 73, as determined from the previously determined distance ratio DC: DB and the distance DB measured in image 101.

In this example, the two reference circles 110 and 114 intersect at two points 76 and 77, one of which is the target structure. A fourth reference point is needed if the proper location of the target cannot be derived from the intersection of the two. Step 136 demonstrates the use of a third set of triangular parameters based on the fourth reference point 74.

The third set of triangle parameters will include the distance ratio FC: FB corresponding to the third triangle FCB and the angle 3. The third set of triangular parameters is used to generate the third reference circle 116. A circle 116 is generated by rotating an orbit extending from the fourth reference point 74 at an angle of 3 with respect to the side FB of the triangle FCB (extending between the reference points 74 and 72). The circumference of the circle 116 is located at a distance FC from the reference point 74, as determined from the measured distance FB in image 101 and the previously determined distance ratio FC: FB. Between the three reference circles 110, 114, and 116, there will be one common intersection 76, which defines the target structure location. The plotted points 76 are displayed on the intraoperative image and provide a target for the clinician to guide the instrument during the invasive procedure.

FIG. 4 is a flow chart summarizing the steps involved in the method for taking triangular parameter measurements for use in a two-dimensional image-guided surgical procedure. During the intervention, intraoperative images may be acquired in 2D or 3D. If 3D intraoperative imaging is available, the triangular parameters obtained using method 10 in FIG. 1 may be applied directly in the intraoperative 3D image, as described in connection with FIG. 3 above. However, if intraoperative imaging is limited to 2D images such as 2D X-ray fluoroscopy, the 3D landmark coordinates acquired during the data acquisition method will be on the plane view selected to calculate the triangular parameters. Need to be projected on.

Method 10'shown in FIG. 4 involves steps 15-35, as previously mentioned in connection with FIG. 1, to obtain landmark point coordinates for each of the target structure and the reference structure. In 2D image-guided applications, it is expected that the two reference structures are generally sufficient to be used when plotting estimated target structure locations. Before calculating the triangular parameters in step 60, the landmark points need to be projected onto the 2D plane view that will be used during the planned procedure.

In step 40 of method 10', one location is calculated within the 3D image volume, and this location is equal to the isocenter axis of the patient's body or body region, where the 2D image rotates around. For example, in a general invasive procedure guided by 2D fluoroscopy, the isocenter axis of the thoracic cavity where the fluoroscopy C-arm will rotate around may be determined. The isocenter of the body region may be determined by measuring the height and width of the region within the imaging volume at the specified location. The axis that passes through the center point of each dimension is determined as the isocenter axis. Other methods may be used to determine the location of the isocenter points used to define the isocenter axis around which the 2D imaging plane rotates.

In one example, the isocenter of the thoracic cavity can be defined by measuring the width and height of the thoracic cavity of each imaging volume, approximately at the height of the aortic valve. The central point of each dimension of the thoracic cavity is then determined. A vertical axis defined through the body at the central intersection of each dimension is used as the axis of rotation on which each set of landmark points can rotate around.

FIG. 5 shows the isocenter 80 of the patient's thoracic cavity and the isocenter axis 82, which allows a standard C-arm to rotate around to obtain a 2D fluoroscopic view. Several angles of rotation used for fluoroscopic views are shown: 30 ° right front tilt angle 84, 30 ° left front tilt angle 86, and 60 ° left front tilt angle 88.

In step 45 of the flowchart of FIG. 4, the landmark point coordinates of the reference structure and the target structure are rotated around the isocenter axis in order to project the landmark points onto the selected planar view. The selected planar view corresponds to a planar view intended to be used during a surgical procedure for image guidance. Another Matlab code developed by the present inventor and called "CSlocrotation.m" can be used in step 45 to integrate each of the three landmark point coordinates with the isocenter defined point, allowing the user to integrate. However, it allows the landmark points to be rotated in either direction at various angles around the isocenter axis.

After these points have been rotated to a given angle around the isocenter axis, in step 50, the points are projected onto the selected plane using the parallel projection method. The resulting 2D landmark point coordinates are then plotted to represent how they appear on the 2D image at a given angular rotation. The triangle parameters calculated in step 60 are the distance ratio and angle obtained from the triangle formed by the projected landmark point coordinates in the selected 2D plane view.

Figure 6 shows the triangular parameters that can be calculated from the projected landmark point coordinates in a 2D system. When the landmark points mentioned in the previous example for localizing CSos are rotated around the isocenter of the thoracic cavity to a left anterior tilt (LAO) angle of 30 °, the projected points are shown in Figure 6. It looks like a triangle. In Figure 6, two projected reference landmark points 90 and 92, labeled A and B, respectively, and a projected target structure land, labeled C, on the selected image plane 95. The mark points 94 form a triangle ABC. In the example of localizing CSos, the reference landmark point 90 corresponds to the superior vena cava landmark point, the reference landmark point 92 corresponds to the tricuspid valve landmark point 92, and the target landmark point 94 corresponds to CSos. obtain.

The distances AB, BC, and AC of each hypotenuse of the triangle ABC formed by the projected landmark points 90, 92, and 94 can be calculated from the projected landmark coordinate values. These distances are used to calculate the distance ratio that associates the lengths of the sides of the triangle ABC.

Normally, two distances are calculated, including 1) the distance from the first reference landmark point to the second reference landmark point and 2) the distance from the first reference landmark point to the target landmark point. become. In the example shown, the distance AB, which is the distance between the reference landmark 90 and the reference landmark 92, can be calculated together with the distance AC, which is the distance between the reference landmark 90 and the target landmark 94. .. The ratio of distance AB to distance AC may then be calculated. In another example, the distance AB and the distance BC may be calculated to determine the distance ratio. The distance ratio generally associates the distance between two reference landmark points 90 and 92 with the distance between one of the reference landmark points 90 or 92 and the target landmark point 94.

In addition to the distance ratio, at least one angle is measured. When the distance ratio is calculated using the distance AB and the distance AC, the angle 1 formed between the vector AB and the vector AC is calculated from the coordinates of each landmark point. When the distance ratio is calculated using the distance CA and the distance CB, the angle 3 formed by these vectors is calculated. By knowing the distance ratio and associated angle measurements, the location of the target structure can be predicted in the intraoperative image containing the reference structure.

FIG. 7 is a flow diagram showing the steps involved in applying triangular parameters during an image-guided procedure to plot the target structure on a two-dimensional intraoperative image. At step 122, the location of reference structure 90 and reference structure 92 is marked by the user within the intraoperative image frame 121. Image frame 121 is a 2D image of the selected view plane. The user can specify the location of reference structure 90 and reference structure 92 using an instruction tool, touch screen, or other user equipment compatible with the imaging system. The user should attempt to mark the location of each reference structure at the point corresponding to each landmark point used when retrieving the triangular data, in order to improve the accuracy of the plotted target structure locations.

After marking the reference structure location, in step 124, the distance between the reference structure 90 and the reference structure 92 in the intraoperative image 121 is measured. The measured distance corresponds to the distance AB of one hypotenuse of the triangle formed by the two reference structures and the target structure, as shown in FIG. Distance AB can be measured in any unit, as a unitless distance ratio will be applied to estimate the location of the target structure.

At step 126, triangular parameters are applied to plot the estimated location 96 of the target structure. In one embodiment, using the labeling arrangement shown in FIG. 6, the measured distance AB at image frame 121 is multiplied by the distance ratio AC: AB to the distance AC between the reference structure 90 and the target structure. Is required. Point 96 is then plotted at a distance AC from reference structure 90 along a vector extending from side AB at an angle of 1. The plotted points 96 provide a target for the clinician to guide the instrument during the invasive procedure. The area or perimeter surrounding the plotted points 96 may be shaded or contoured to indicate the inherent error or variation of the triangular parameters.

The above-mentioned example for localizing CSos may be used in clinical situations to further describe how to apply the mean triangular parameter. Triangular parameters associating the target CSos as well as the location of SVC and TV reference structures can be obtained from a large number of subjects from selected patient populations. From a given patient population, in the LAO 30 ° observation direction (as shown in Figure 5), the distance ratio between the SVC-TV hypotenuse and the SVC-CSos hypotenuse and the SVC-TV hypotenuse and the SVC-CSos hypotenuse of the triangle. The assumption may be made that the variation in angle between them is small.

During a procedure requiring CSos cannulation, the doctor controls the orientation of the fluoroscopy C-arm to observe the patient's heart from a LAO 30 rotation angle, and then using a catheter, the doctor Position the center points of the SVC port and TV wheel and mark these locations on the fluoroscopic view. These points may be manually marked on the monitor displaying the fluoroscopic image, or electronically marked on the image by the software package, as well as on the fluoroscopic view. The distance between these points can be measured using a ruler and a line drawn between the two points, or can be calculated by a software package. Then, using the previously generated triangular parameter database for the selected patient population, the length of the SVC-CSos hypotenuse of the triangle is the distance ratio between the SVC-TV hypotenuse and the SVC-CSos hypotenuse of the triangle. It will be calculated using the measured values. The orbit of the triangular SVC-CSos hypotenuse is then defined by using the existing line (SVC-TV hypotenuse) on the fluoroscopic view and the angle measurements already defined from the triangle parameter database. It will be.

After plotting on the X-ray fluoroscopy screen the points corresponding to the expected location of the CSos, obtained from the length and orbital calculations, the variation-based perimeter found in the population is around the expected CSos location. Draw and provide a target to guide the physician to cannulate into the coronary sinus. When patient-specific localization methods are used, physicians can follow the same steps as the steps outlined above for population-based tools, the only difference being smaller targets because there is no variation in the patient population. That is.

FIG. 8 summarizes the steps contained in yet another embodiment of the invention in which localization parameters derived from one reference structure and local coordinate system are used when plotting target structure locations within intraoperative images. It is a figure. For some applications, a single reference structure 90 may be selected. The data collection method performed prior to the surgical procedure measures the distance between the reference structure landmark point and the target landmark point. The angle between the local coordinate axis and the trajectory extending from the reference landmark point in the direction of the target landmark point is obtained. The local coordinate axes are defined for the imaging area 121. This distance measurement and angle are used in the method of plotting the estimated target structure location during the image guidance procedure.

As shown in FIG. 8, the location of the reference structure 90 within the 2D intraoperative image 101 is marked by the clinician at step 105. The axis of the local coordinate system, in this example, an orbit extending from the reference point 90 at an angle of 1'from the Y axis is generated. At step 107, the target point 96 is plotted along the trajectory at a distance AC from the reference point 90. Distance AC corresponds to the distance measured during data acquisition between the reference point 90 and the target landmark point. Therefore, in some applications, target point locations may be plotted using at least one reference structure and a defined local coordinate system. The localization parameter data includes absolute distance measurements rather than distance ratios.

The methods described herein may be applied in a stepwise manner to position a series of structures. For example, in procedures where catheters or leads are placed intracardiac veins, the methods described herein may first be applied to localize the coronary sinus ostium. Once the coronary sinus ostium is positioned, the coronary venous sinus ostium may be used as a reference structure for positioning the target cardiac vein. To localize the target cardiac vein, the location of the CSos and a second reference structure, eg, a triangular parameter that associates the tricuspid valve, may be obtained. Thus, the target structure in the first localization step may be used as a reference structure in the second localization step performed to finally guide the instrument to the final treatment delivery site or diagnostic site.

Thus, a method of plotting the target structure location on the intraoperative image based on the previously obtained triangular parameters was described. The methods provided by the present invention can be used in any intervention that requires precise structural localization using a 2D or 3D imaging system. Procedures in which the present invention may be beneficial include CSos cannula insertion, pacing lead placement, local delivery of drugs, biological or genetic material, tissue resection, percutaneous valve placement, and physiological sensors. Including, but not limited to, installation. By using the present invention to assist clinicians in localizing target structures, radiation exposure can be reduced by reducing procedure time and requiring only a small number of fluoroscopic images. It provides more predictable treatment results. The use of localization methods increases physician confidence when performing a particular procedure and reduces the physician training time required to learn the procedure. The methods described herein are intended to provide exemplary methods for carrying out the present invention and are not intended to limit the scope of the appended claims.

<figref num="1">It is a flowchart summarizing the steps included in the data collection method of calculating the triangular parameter data which associates a 3D reference structure with a target structure.</figref><figref num="2A">It is a figure which shows the set of triangular parameters which can be calculated in the data collection method shown in FIG.</figref><figref num="2B">It is a figure which shows the set of triangular parameters which can be calculated in the data collection method shown in FIG.</figref><figref num="3A">It is a diagram of the steps involved in the method of applying 3D triangular parameters when plotting the estimated target structure location in a 3D intraoperative image.</figref><figref num="3B">FIG. 5 illustrates the use of four reference landmark points to plot estimated target structure locations within a 3D intraoperative image.</figref><figref num="4">FIG. 6 is a flow chart summarizing the steps included in the method for collecting triangular parameter measurements for use in a two-dimensional image-guided surgical procedure.</figref><figref num="5">FIG. 5 shows an isocenter of the patient's thoracic cavity and an isocenter axis on which a standard C-arm can rotate around to obtain a 2D fluoroscopic view.</figref><figref num="6">It is a figure which shows the triangle parameter which can be calculated from the landmark point coordinates projected in a 2D system.</figref><figref num="7">FIG. 6 is a flow diagram showing the steps involved in applying triangular parameters during an image-guided procedure to plot the target structure on a two-dimensional intraoperative image.</figref><figref num="8">It is a diagram summarizing the steps included in yet another embodiment of the present invention, in which one reference structure and localization parameters obtained from the local coordinate system are used when plotting the target structure location in the intraoperative image.</figref>