JP7246907B2 - Scoring of myocardial nuclear medicine image data - Google Patents

Description

The present application relates to scoring nuclear myocardial medicine image data.

Nuclear medicine techniques are often used to study myocardial function. One such technique is myocardial scintigraphy, which is a technique for imaging the uptake of intravenously administered radiopharmaceuticals into the myocardium by detecting the radiation emitted from the radiopharmaceuticals. While CT, which uses X-rays, aims to obtain information on the shape of tissues such as blood vessels, myocardial scintigraphy examines the functions of the myocardium, such as the amount of blood flowing into the myocardium and energy metabolism. for the purpose of

As a method for evaluating myocardial scintigrams, conventionally, scoring based on radiation count values (hereinafter referred to as scoring) has been used. The radiation count value reflects the amount of radiopharmaceutical uptake. In this method, the myocardium is divided into several regions, and scoring is performed for each region to provide information that serves as a reference for evaluation in many cases. As a method of segmentation, a model that simply divides into three regions, the left anterior descending artery region, the left circumflex artery region, and the right coronary artery region, is commonly used by the Japanese Society of Nuclear Medicine, the Japanese Society of Nuclear Medicine, and the American Heart Association. A 17-segment model is often used.

Further, Patent Literature 1 listed below describes specifying a region governed by a coronary artery based on a morphological image of a coronary artery obtained by CT or MRI, and analyzing a SPECT image for each region governed by the coronary artery.

JP 2011-025005 A

However, the clinical importance of myocardial regions varies from region to region. In other words, even if the scoring results are the same in two areas, their clinical significance may differ.

Therefore, the present application discloses an invention for solving the above-described problems, and an invention for scoring myocardial nuclear medicine image data. This invention
extracting coronary arteries and left ventricular myocardium from coronary artery angiogram image data, and dividing the extracted left ventricular myocardium based on the area covered by the extracted coronary arteries;
calculating the volume of each divided myocardial region;
performing radiation count-based scoring for each of the divided myocardial regions based on the coronary artery angiogram image data and myocardial nuclear medicine image data co-registered;
for each myocardial region, correcting the scoring results based on the volume;
including.

By correcting the scoring results of a myocardial region based on the volume of that region, the scoring results reflect the importance of each region. Therefore, the scoring results can be clinically more useful.

For example, the Voronoi method may be used to divide the extracted left ventricular myocardium based on the extracted coronary artery control area.

By using the Voronoi method, it is possible to perform region segmentation by reflecting individual differences in the course of coronary arteries. Therefore, appropriate area division can be performed for each image data to be analyzed. By performing the above-described volume correction on segmentation using the Voronoi method, the results of scoring myocardial nuclear medicine image data can be clinically more useful. For example, it may be possible to identify areas of higher treatment priority.

In a preferred embodiment, the correcting comprises
calculating, for each myocardial region, the ratio of the volume of the myocardial region to the volume of the entire left ventricle;
multiplying the scoring result by the percentage for each of the myocardial regions;
may include

An example of an embodiment of the invention is a method performed by a device by means of processing means of the device executing program instructions, comprising:
extracting coronary arteries and left ventricular myocardium from coronary artery angiogram image data, and dividing the extracted left ventricular myocardium based on the area covered by the extracted coronary arteries;
calculating the volume of each divided myocardial region;
performing radiation count-based scoring for each of the divided myocardial regions based on the coronary artery angiogram image data and myocardial nuclear medicine image data co-registered;
for each myocardial region, correcting the scoring results based on the volume;
A method is included, comprising:

Another example of an embodiment of the invention comprises a computer program comprising program instructions which, when executed by processing means of an apparatus, cause said apparatus to perform the method described above.
Another example of an embodiment of the invention is an apparatus comprising processing means and storage means, wherein program instructions are stored in said storage means, said program instructions, when executed by said processing means, said apparatus comprising: includes an apparatus configured to cause the method to be performed.

Another example of an embodiment of the invention includes:
means for extracting coronary arteries and left ventricular myocardium from coronary artery angiogram image data and segmenting the extracted left ventricular myocardium based on the extracted coronary artery control area;
means for calculating the volume of each divided myocardial region;
means for performing radiation count-based scoring for each of the divided myocardial regions based on the coronary artery angiogram image data and myocardial nuclear medicine image data co-registered;
means for correcting the scoring result based on the volume for each myocardial region;
includes a device comprising

In some embodiments, the coronary artery angiogram image data can be image data acquired using CT. In some embodiments, the coronary artery angiogram image data can be image data acquired using MRI.
In some embodiments, the nuclear myocardial medicine image data can be image data acquired by SPECT. In some embodiments, the nuclear myocardial medicine image data can be image data acquired by PET.

Some of the presently preferred implementations of the invention are specified in the claims included below. However, the configurations specified in these claims do not necessarily include all of the novel technical ideas disclosed in the specification and drawings of the present application. Applicant claims to have the right to obtain a patent for all novel technical ideas disclosed in the specification and drawings, whether or not they are included in the present claims. write down

It is a figure for demonstrating the hardware constitutions of the system which can implement this invention. An example of processing for scoring myocardial scintigram data is shown. FIG. 12 shows an example of segmentation of the left ventricular myocardium based on coronary tributaries. FIG. An example of scoring results of myocardial scintigram data is shown.

Hereinafter, preferred embodiments of the technical idea disclosed in the present application will be described with reference to the accompanying drawings.

FIG. 1 is a diagram for explaining the hardware configuration of a system 100 that can implement the present invention. As depicted in FIG. 1, the system 100 is similar in hardware to a typical computer and includes a CPU 102, a main memory 104, a mass storage device 106, a display interface 107, a peripheral device interface 108, and a network. • An interface 109 or the like may be provided. As with a general computer, a high-speed RAM (random access memory) can be used as the main memory device 104, and a low-cost, large-capacity hard disk or SSD can be used as the large-capacity storage device 106. can. A display can be connected to system 100 for displaying information, and is connected via display interface 107 . A user interface such as a keyboard, mouse, or touch panel can also be connected to the system 100 via a peripheral interface 108 . Network interface 109 can be used to connect to other computers or the Internet over a network.

The mass storage device 106 stores an operating system (OS) 110, a coregistration program 120, a segmentation/volume calculation program 122, and a scoring/weighting program 124. FIG. The most basic functions of system 100 are provided by OS 110 being executed by CPU 102 . The co-registration program 120 is a program for co-registration of CT image data 130 and nuclear medicine image data 132 . The segmentation/volume calculation program 122 has program instructions for the novel processing disclosed by the present application. At least part of the instructions are executed by the CPU 102, whereby the system 100 can automatically segment the CT image data (the coronary artery CT angiogram image data 130), that is, divide the regions. The region division/volume calculation program 122 can further calculate the volume of each region and the volume ratio to the whole.

Scoring and weighting program 124 also includes program instructions for the novel processes disclosed by this application. At least part of the instructions are executed by the CPU 102, so that the system 100 refers to the myocardial scintigram image data 132 for each of the myocardial regions divided by the program 122, and performs scoring based on the radiation count value. conduct. The scoring/weighting program 124 further corrects the scoring result by multiplying the scoring result by the volume ratio calculated by the region dividing/volume calculating program 122 .

Mass storage device 106 may also store CT image data 130 and nuclear medicine image data 132 . CT image data 130 is image data obtained from a coronary CT angiogram. A coronary artery CT angiogram is a technique for imaging coronary arteries by CT using a contrast medium, and is a technique often used for examining coronary artery stenosis and ischemic heart disease. The CT image data 130 is three-dimensional image data in which each pixel value corresponds to a CT value, and is subject to analysis or manipulation by the segmentation/volume calculation program 122 .

The nuclear medicine image data 132 is image data obtained by myocardial scintigram. A myocardial scintigram is a technique based on nuclear medicine, and is a technique for imaging the state in which a radioactive drug administered through a vein is taken into the myocardium by detecting the radiation emitted from the radioactive drug. As a nuclear medicine technique, SPECT is usually used, but PET is sometimes used. Radiopharmaceuticals for SPECT include thallium chloride ( ²⁰¹ TlCl) injection, tetrofosmin technetium ( ^99m Tc Tetrofosmin) injection, hexakis(2-methoxyisobutylisonitrile) technetium ( ^99m Tc MIBI) injection, 15-( There are 4-iodophenyl)-3(R,S)-methylpentadecanoic acid ( ¹²³ I) injections, and 13N-ammonia and rubidium chloride ( ⁸² RbCl) injections for PET preparations. The image data 132 in this example is image data obtained by SPECT. The myocardial scintigram image data 132 is three-dimensional image data in which each pixel value corresponds to a radiation count value, and is subject to analysis or manipulation by the scoring/weighting program 124 .

The large-capacity storage device 106 further contains area data 140 for specifying each area specified by the area division/volume calculation program 122, and the volume and volume ratio of each area calculated by the area division/volume calculation program 122. Data 142, score data 144 calculated by the scoring and weighting program 124, and the like may be stored.

In addition to the elements depicted in FIG. 1, the system 100 can have a configuration similar to that of a typical computer system, such as a power supply and a cooling device. Implementation of computer systems includes distributed/redundant and virtualized storage devices, use of multiple CPUs, CPU virtualization, use of processors specialized for specific processing such as DSPs, and hardware implementation of decision processing in CPUs. Various forms using various techniques such as combining are known. The invention disclosed in the present application may be installed on any form of computer system, and its scope is not limited by the form of the computer system. The technical ideas disclosed in this specification are generally: (1) executed by a processing means to cause a device or system including the processing means to perform various processes described in this specification; (2) a method of operating a device or system realized by said processing means executing said program; (3) said program and said program configured to execute said program It can be embodied as a device, system or the like comprising processing means. As mentioned above, part of the software processing may be implemented in hardware.

Also, it should be noted that the CT image data 130 and the SPECT image data 132 are often not stored in the mass storage device 106 when the system 100 is manufactured, sold, or activated. Data 132 may be data transferred to system 100 from an external device via peripheral interface 108 or network interface 109, for example. The data 140, 142, 144 may be data stored in the mass storage device 106 as a result of at least part of the segmentation/volume calculation program 122 and the scoring/weighting program 124 being executed by the CPU 102. . It should be noted that the scope of the invention disclosed in the present application is not limited by whether image data or the like is stored in the storage device.

Next, FIG. 2 is used to describe the flow of process 200, which is an example of the novel scoring process disclosed in the present application. The process 200 can be, for example, a process performed by the system 100 by causing the CPU 102 to execute the segmentation/volume calculation program 122, the scoring/weighting program 124, and the like.

Step 202 marks the beginning of the process. At step 204, co-registration of the coronary CT angiogram image data 130 and the myocardial scintigram image data 132 is performed. That is, the orientation, size, and position of the body are three-dimensionally matched between the coronary artery CT angiogram image data 130 and the myocardial scintigram image data 132 . By performing this process, it becomes easy to use these two image data. For example, it is possible to display the myocardial scintigram image data 132 superimposed on the coronary artery CT angiogram image data 130 . The co-registration process may be a process performed by the system 100 as the co-registration program 120 is executed by the CPU 102 .

Since the coregistration function is included in many medical image data analysis programs available on the market, in this example as well, existing techniques implemented in such programs are used to perform coregistration. may be performed. Coregistration can be performed using, for example, the residual sum of squares. For example, one of the CT and SPECT images is rotated or resized, and the pixel-by-pixel difference from the other image is calculated. can do. Coregistration may also be by manual methods. That is, both the CT image 130 and the nuclear medicine image 132 may be displayed as images, and one of the images may be translated or rotated by operating a mouse or the like to be combined with the other image.

In the following discussion, it should be understood that image data 130 and 132 have been co-registered. In some embodiments, coregistration between image data 130 and 132 may already be completed. For example, if image data 130 and 132 were acquired together by an integrated CT and SPECT system, coregistration of coronary CT angiogram image data 130 and myocardial scintigram image data 132 would be , may have already been completed when output from the device. In that case, of course, the process of step 208 is unnecessary. In such embodiments, coregistration program 120 would also not be necessary.

In step 206, the coronary arteries and left ventricular myocardium are extracted from the coronary artery CT angiogram image data 130, and the extracted left ventricular myocardium is segmented based on the regions of control of the extracted coronary arteries.

Many commercially available CT image data analysis programs also include functions for extracting coronary arteries and left ventricular myocardium from coronary CT angiogram image data. In addition, these extraction methods are also described in Patent Document 1 cited above. Also in the present embodiment, myocardial extraction may be performed using such an existing method.

Dividing the coronary artery CT angiogram image data based on the extracted area of control of the coronary artery can be performed using, for example, the Voronoi method. Techniques for segmenting a CT image based on the Voronoi method are already known.
https://www.cg.tuwien.ac.at/research/publications/2009/termeer-2009-scmr/
You can see an example in

Note that in some embodiments, myocardial contour extraction of CT and/or SPECT images may be required to perform co-registration of CT and SPECT images. In such embodiments, step 206 (at least the myocardium extraction process) would precede step 204 .

In some embodiments, the results of region segmentation in step 206 may be stored as region data 140 in mass storage device 106 . For example, the region data 140 is configured and stored as data having the same element array as the image data 130 and 132 after co-registration, and the value of each element is the region identification value (1, 2, 3, etc.). It is possible to

At step 208, for each of the regions divided at step 206, the volume is calculated. The volume may be calculated based on, for example, the number of pixels included in each divided region in the CT image data 130 .

Also in step 208, for each myocardial region, the ratio of the volume of the myocardial region to the volume of the entire left ventricular myocardium is also calculated. This ratio can be calculated, for example, by dividing the number of pixels contained in each region by the number of pixels in the entire left ventricular myocardium.

In some embodiments, the results of calculating the volumes and proportions in step 208 may be stored as volume and proportion data 142 in mass storage device 106 .

At step 210, for each of the regions divided at step 206, the count value of the corresponding region in the myocardial scintigram image data 132 is examined, and each region is scored based on the count value. Although various scoring methods may be used, for example, the following method may be used.

(1) For each region, convert all pixel values in the region to a ratio of the maximum pixel value in the region. For example, it may be converted to a percentage, and the converted value is called %uptake.
(2) Calculate the average value of %uptake for each region.
(3) Assign points (scores) to the calculated average %uptake value according to a preset threshold. For example, 0 points for 100% to 80%, 1 point for 80% to 70%, 2 points for 70% to 60%, 3 points for 60% to 50%, 4 points for less than 50%, etc. . Depending on the embodiment, these thresholds may be arbitrarily set by the user, or may be set based on the average value and degree of variation in the normal group.

Other known scoring methods such as Subtraction Score, Mismatch Score, Severity Score, Extent Score, etc. may be used. In addition, instead of converting each pixel value into a ratio to the maximum pixel value in the region, the washout rate, that is, the signal intensity during the rest test subtracted from the signal intensity (count value / pixel value) during the stress test, is used. It may be converted to a value divided by the signal strength under load. Then, scoring may be performed based on the Washout rate.

In some embodiments, scoring may be performed manually by the user. For example, scoring may be performed in consideration of the color tone of an image obtained by developing the left ventricular myocardium in polar coordinates and the degree of spread of low pixel values in the region. Manual scoring has the advantage of facilitating artifact removal. For example, in automatic processing using a threshold, if there is an extremely low pixel value localized to a very small part of the region, the average value of the entire region will be lowered due to the pixel value, and the score will be high. However, if scoring is performed visually, measures can be easily taken to exclude such extreme pixel values (regions) from the evaluation of the score.

In some embodiments, this scoring may be performed on a region-by-region basis using all pixels contained in that region. Depending on the embodiment, this may be done using only some of the pixels in the area.

In some embodiments, the rest test score may be subtracted from the stress test score for each region.

In some embodiments, the results of scoring in step 210 may be stored as score data 144 on mass storage device 106 .

At step 212, each segmented region (at step 206) based on the coronary artery coverage area is weighted by multiplying the scoring result at step 210 by the volume fraction calculated at step 208.

In some embodiments, the weighted scoring results of step 212 may be stored in mass storage device 106 as score data 144 .

At step 214, the weighted scoring results are displayed. A display example will be described later with reference to FIGS.

Step 216 indicates the end of processing.

FIG. 3 illustrates the segmentation of the left ventricular myocardium performed in step 206 . FIG. 3 depicts a two-dimensional unfolded view of the left ventricular myocardium, divided into LAD (left anterior descending artery), LCX (left circumflex artery), and RCA (right coronary artery) innervating regions. I know there is. Diag (diagonal branch), which is often treated as a part of LAD, is also divided into its dominant regions.

Also shown in FIG. 3 are the results of the volume calculation for each region, performed in step 208 .

FIG. 4 shows an example of the results of scoring performed in steps 210 and 212 and their display, comparing them with the results of the conventional method. Six graphs are drawn in FIG. 4, each of which is obtained by dividing a two-dimensional unfolded view of the left ventricular myocardium based on the coronary artery control region and plotting the results of scoring. Since FIGS. 3 and 4 are obtained based on CT images and SPECT images continuously acquired from the same subject, the state of segmentation is common between FIGS. 3 and 4. FIG.

The upper three in FIG. 4 show the results of the conventional method, and the lower three show the results of the present invention. In addition, the leftmost column shows the score obtained from the stress measurement, the middle column shows the score obtained from the resting measurement, and the rightmost column shows the value obtained by subtracting the resting score from the stress score. ing. In this example, each score is based on SPECT data using 99mTc-tetrofosmin as the radiopharmaceutical. The higher the score, the lower the amount of radiopharmaceutical uptake, that is, the higher the possibility of coronary artery stenosis and myocardial function deterioration.

Comparing the top and bottom graphs in the leftmost column, without volume correction, the coronary artery-dominated region with the highest score during stress is the LCX-dominated region. That is, the region that should be most clinically noted is the LCX-dominated region. However, with volume correction, the coronary artery-dominated region with the highest score during stress, ie, the region that deserves the most clinical attention, is the LAD-dominated region.

While the invention has been described in detail with reference to preferred embodiments, the foregoing description and accompanying drawings are not intended to limit the scope of the invention, but rather to satisfy legal requirements. was presented in Embodiments of the present invention have various variations other than those introduced here. For example, in some embodiments, MRI data may be used as coronary artery angiogram image data instead of CT data used in the above example. Further, depending on the embodiment, data acquired using PET may be used as myocardial nuclear medicine image data instead of the SPECT data used in the above example. Various numerical values shown in the specification or drawings are also examples, and these numerical values are not presented with the intention of limiting the scope of the invention. Individual features included in the various embodiments shown in the specification or drawings are not intended to be used only in conjunction with the embodiments in which that feature is directly recited, but rather in conjunction with other embodiments described herein. It is also possible to use them in combination in the examples and various implementation examples not described. In particular, the order of the processes introduced in the flow charts does not necessarily have to be executed in the order introduced, and according to the preferences and needs of the implementer, the order can be changed, they can be executed in parallel, or multiple processes can be executed. may be implemented such that the block of is implemented indivisibly or executed as an appropriate loop. All of these variations are included in the scope of the invention disclosed in this application, and the scope of the invention is not limited by the processing implementation. The order in which the processes are described in the claims does not necessarily determine the order of the processes. is also included in the scope of the claimed invention.

Further, for example, the region division/volume calculation program 122 and the scoring/weighting program 124 may be a single program or a program group composed of a plurality of independent programs. can be included. These two programs may be provided as a single program in some embodiments. In some embodiments, coregistration program 120 may also be provided. As is well known, there are various implementation forms of programs, and all such variations are included within the scope of the invention disclosed herein. Applicant asserts that it is entitled to a patent for all forms that do not depart from the spirit of the invention disclosed in this application, whether or not claimed in the present claims. Note that

100 system 102 CPU
104 Main storage device 106 Mass storage device 107 Display interface 108 Peripheral device interface 109 Network interface 120 Coregistration program 122 Segmentation/volume calculation program 130 Coronary artery CT angiogram image data 132 Myocardial scintigram image data

Claims (6)

A method performed by a device by means of processing means of the device executing program instructions, comprising:
extracting coronary arteries and left ventricular myocardium from coronary artery angiogram image data, and dividing the extracted left ventricular myocardium based on the area covered by the extracted coronary arteries;
calculating the volume of each divided myocardial region;
performing radiation count-based scoring for each of the divided myocardial regions based on the coronary artery angiogram image data and myocardial nuclear medicine image data co-registered;
for each myocardial region, correcting the scoring results based on the volume;
A method, including 2. The method of claim 1, wherein segmenting the extracted left ventricular myocardium based on the extracted coronary artery coverage is performed using a Voronoi method. The correcting is
calculating, for each myocardial region, the ratio of the volume of the myocardial region to the volume of the entire left ventricular myocardium;
multiplying the scoring result by the percentage for each of the myocardial regions;
3. The method of claim 1 or 2, comprising: 4. A device comprising processing means and storage means, said storage means comprising program instructions which, when executed by said processing means, cause said device to store the data of any of claims 1 to 3. An apparatus configured to carry out the method of A computer program product comprising program instructions adapted to, when executed by means of an apparatus, cause said apparatus to perform the method according to any one of claims 1 to 3. means for extracting coronary arteries and left ventricular myocardium from coronary artery angiogram image data and segmenting the extracted left ventricular myocardium based on the extracted coronary artery control area;
means for calculating the volume of each divided myocardial region;
means for performing radiation count-based scoring for each of the divided myocardial regions based on the coronary artery angiogram image data and myocardial nuclear medicine image data co-registered;
means for correcting the scoring result based on the volume for each myocardial region;
A device comprising:

Images