Myocardial Hypertrophy Measurement Device, System and Storage Medium

Abstract

The present invention belongs to the technical field of image processing, and discloses a myocardial hypertrophy measurement device, system and storage medium. The device includes: an interface configured to receive coronary CTA images of the heart; a processor configured to: based on the coronary CTA images, respectively obtain a myocardial segmentation binary map and a heart segmentation map; based on the heart segmentation map, determine the centroid of the left atrioventricular junction position in the left atrioventricular junction position binary map and the centroid of the right atrioventricular junction position in the right atrioventricular junction position binary map; obtain a short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial annular segmentation map; based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position and the maximum distance point, determine a long-axis plane; based on the long-axis plane, intercept the short-axis myocardial annular segmentation map to obtain a target connected domain; take the maximum distance between any two points in the target connected domain as the measurement value of the myocardial thickness at the middle part. It can quickly and accurately measure the myocardial thickness.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular, to a myocardial hypertrophy measurement device, system and storage medium. Background Art

[0002] Cardiovascular diseases have become one of the main diseases threatening human life safety. Doctors use vascular imaging technology to assist in diagnosing various vascular diseases. In clinical practice, when doctors analyze medical images of the cardiovascular structure, they need to obtain some quantitative results to determine the corresponding treatment plan. Automated measurement and diagnosis can improve the diagnosis efficiency and reduce a large number of repetitive operations of doctors. However, for coronary CTA image data, doctors need to manually measure the myocardial thickness at the atrioventricular junction to diagnose whether a patient has myocardial hypertrophy. This process is cumbersome and time-consuming. Summary of the Invention

[0003] The main purpose of the present invention is to provide a myocardial hypertrophy measurement device, system and storage medium, aiming to solve the technical problem that doctors in the prior art need to manually measure the myocardial thickness of patients to diagnose whether the patients have myocardial hypertrophy.

[0004] To achieve the above object, the present invention provides a myocardial hypertrophy measurement device, and the myocardial hypertrophy diagnosis device includes:

[0005] An interface configured to receive coronary CTA images of the heart;

[0006] A processor configured to:

[0007] Based on the coronary CTA images, respectively obtain a myocardial segmentation binary map of the heart and a heart segmentation map of the heart;

[0008] Based on the heart segmentation map, determine a left atrioventricular junction position binary map and a right atrioventricular junction position binary map;

[0009] Determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map;

[0010] Determine the maximum distance point on the myocardium to the centroid of the left atrioventricular junction position, and determine the center point of the centroid of the left atrioventricular junction position and the maximum distance point, wherein the maximum distance point is determined as the apex position;

[0011] Obtain a short-axis plane, and based on the short-axis plane, intercept the myocardium to obtain a short-axis myocardial annular segmentation map, wherein the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point;

[0012] Determine the long-axis plane based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the apex position;

[0013] Intercept the short-axis myocardial annular segmentation map based on the long-axis plane to obtain a target connected component;

[0014] Determine the positions of two endpoints in the target connected component, and determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the myocardial thickness measurement value at the middle part.

[0015] Optionally, the cardiac segmentation map includes the left ventricle, the right ventricle, the left atrium, and the right atrium; wherein,

[0016] The processor is further configured to:

[0017] Dilate the left ventricle and the left atrium respectively. After the intersection of the dilated left ventricle and the dilated left atrium is not an empty set, determine the binary map of the left atrioventricular junction position based on the intersection;

[0018] Dilate the right ventricle and the right atrium respectively. After the intersection of the dilated right ventricle and the dilated right atrium is not an empty set, determine the binary map of the right atrioventricular junction position based on the intersection.

[0019] Optionally, the processor is further configured to:

[0020] Determine the centroid of the right ventricle according to the cardiac segmentation map;

[0021] Intercept the short-axis myocardial annular segmentation map based on the long-axis plane to obtain a first connected component and a second connected component;

[0022] Determine a first distance from the centroid of the right ventricle to the midpoint of the first connected component, and determine a second distance from the centroid of the right ventricle to the midpoint of the second connected component;

[0023] Determine the smaller distance between the first distance and the second distance, and use the connected component corresponding to the smaller distance as the target connected component.

[0024] Optionally, the processor is further configured to:

[0025] Based on the linear feature of a single pixel in the target connected component, determine the number of pixels occupied by the maximum distance between any two points in the target connected component;

[0026] Determine the myocardial thickness measurement value according to the occupied number of pixels.

[0027] Optionally, the processor is further configured to:

[0028] Input the coronary CTA image into the trained first deep convolutional network to segment the myocardium, and obtain the binary myocardium segmentation map of the heart;

[0029] Input the coronary CTA image into the trained second deep convolutional network to segment the heart, and obtain the heart segmentation map of the heart.

[0030] Optionally, the processor is further configured to:

[0031] Determine the myocardial hypertrophy diagnosis threshold;

[0032] Judge whether the measured value of the myocardial thickness at the middle is greater than the myocardial hypertrophy diagnosis threshold;

[0033] If it is determined that the measured value of the myocardial thickness at the middle is greater than the myocardial hypertrophy diagnosis threshold, diagnose the patient corresponding to the heart as a patient with myocardial hypertrophy.

[0034] Optionally, the processor is further configured to:

[0035] Determine the first target point on the line segment O between the centroid of the left atrioventricular junction and the maximum distance point, where the distance between the first target point and the centroid of the left atrioventricular junction is between 0.2 - 0.3O 左房室 O 心尖 ; 左房室 O 心尖

[0036] Obtain the first short-axis plane, and intercept the myocardium based on the first short-axis plane to obtain the first short-axis myocardial annular segmentation map, where the first short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction and the maximum distance point and passing through the first target point;

[0037] Based on the long-axis plane, intercept the first short-axis myocardial annular segmentation map to obtain the target connected region;

[0038] Determine the positions of the two endpoints in the target connected region, and determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the measured value of the myocardial thickness at the base of the heart.

[0039] Optionally, the processor is further configured to:

[0040] Determine the second target point on the line segment O between the centroid of the left atrioventricular junction and the maximum distance point, where the distance between the second target point and the centroid of the left atrioventricular junction is between 0.7 - 0.8O 左房室 O 心尖 ; 左房室 O 心尖 ​​

[0041] Obtain the second short-axis plane, and intercept the myocardium based on the second short-axis plane to obtain a second short-axis myocardial annular segmentation map, where the second short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the second target point;

[0042] Based on the long-axis plane, intercept the second short-axis myocardial annular segmentation map to obtain a target connected region;

[0043] Perform a skeletonization process on the target connected region to obtain a skeletonized target connected region;

[0044] Determine the positions of two endpoints based on the skeletonized target connected region, determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the measurement value of the myocardial thickness at the apex.

[0045] In addition, to achieve the above object, the present invention also proposes a myocardial hypertrophy measurement system, and the myocardial hypertrophy measurement system includes the myocardial hypertrophy measurement device as described above:

[0046] The myocardial hypertrophy diagnosis device includes:

[0047] An interface configured to receive coronary CTA images of the heart;

[0048] A processor configured to:

[0049] Based on the coronary CTA images, respectively obtain the myocardial segmentation binary map of the heart and the heart segmentation map of the heart;

[0050] Based on the heart segmentation map, determine the left atrioventricular junction position binary map and the right atrioventricular junction position binary map;

[0051] Determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map;

[0052] Determine the maximum distance point on the myocardium to the centroid of the left atrioventricular junction position, and determine the center point between the centroid of the left atrioventricular junction position and the maximum distance point, where the maximum distance point is determined as the apex position;

[0053] Obtain the short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial annular segmentation map, where the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point;

[0054] Based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the apex position, determine the long-axis plane;

[0055] Based on the long-axis plane, intercept the short-axis myocardial circular segmentation map to obtain a target connected region;

[0056] Determine the positions of two endpoints in the target connected region, and determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the measured value of the myocardial thickness at the middle part.

[0057] In addition, to achieve the above object, the present invention also proposes a storage medium, on which a myocardial hypertrophy measurement program is stored. When the myocardial hypertrophy measurement program is executed by a processor, the following steps are implemented:

[0058] Based on the coronary CTA image, respectively obtain the myocardial segmentation binary map of the heart and the heart segmentation map of the heart;

[0059] Based on the heart segmentation map, determine the left atrioventricular junction position binary map and the right atrioventricular junction position binary map;

[0060] Determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map;

[0061] Determine the point on the myocardium with the maximum distance from the centroid of the left atrioventricular junction position, and determine the center point between the centroid of the left atrioventricular junction position and the maximum distance point;

[0062] Obtain the short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial circular segmentation map, where the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the center point;

[0063] Based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the maximum distance point, determine the long-axis plane;

[0064] Based on the long-axis plane, intercept the short-axis myocardial circular segmentation map to obtain a target connected region;

[0065] Determine that the maximum distance between any two points in the target connected region is the measured value of the myocardial thickness at the middle part.

[0066] The myocardial hypertrophy measurement device, system and storage medium proposed by the present invention. The myocardial hypertrophy diagnosis device includes: an interface configured to receive coronary CTA images of the heart; a processor configured to: based on the coronary CTA images, respectively obtain a myocardial segmentation binary map of the heart and a heart segmentation map of the heart; based on the heart segmentation map, determine a left atrioventricular junction position binary map and a right atrioventricular junction position binary map; determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map; determine the point on the myocardium with the maximum distance to the centroid of the left atrioventricular junction position, and determine the center point between the centroid of the left atrioventricular junction position and the maximum distance point, wherein the maximum distance point is determined as the apex position; obtain a short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial annular segmentation map, wherein the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point; based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position and the apex position, determine a long-axis plane; based on the long-axis plane, intercept the short-axis myocardial annular segmentation map to obtain a target connected region; determine the

[0067] positions of two end points in the target connected region, and determine the distance between the two end points according to the positions of the two end points. 5 Take the distance as the measurement value of the myocardial thickness at the middle part. Through the above method, the myocardial thickness can be measured quickly and accurately.

[0068] Measure the myocardial thickness. Description of the Drawings

[0069] Figure 1 is a schematic structural diagram of a myocardial hypertrophy measurement system in the hardware operating environment related to the embodiment solution of the present invention;

[0070] Figure 2 is a schematic flowchart of the processing method of the first embodiment of the myocardial hypertrophy measurement device of the present invention;

[0071] Figure 3 is a schematic flowchart of the processing method of the first embodiment of the myocardial hypertrophy measurement device of the present invention;

[0072] Figure 4 is a schematic diagram of the long-axis plane intercepting the short-axis myocardial annular segmentation map in the first embodiment of the myocardial hypertrophy measurement device of the present invention;

[0073] shape segmentation map;

[0074] Figure 5 is a schematic flowchart of the processing method of the second embodiment of the myocardial hypertrophy measurement device of the present invention;

[0075] Figure 6Schematic flowchart of the processing method according to the third embodiment of the myocardial hypertrophy measurement device of the present invention.

[0076] The realization of the object, functional characteristics and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0077] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0078] Referring to Figure 1 , Figure 1 Schematic structural diagram of the myocardial hypertrophy measurement 5 system of the hardware operating environment involved in the embodiment solution of the present invention.

[0079] As Figure 1 shown, the myocardial hypertrophy measurement system may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection communication between these components.

[0080] The user interface 1003 may include a display screen (Display) and an input unit such as a keyboard 0 (Keyboard). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless-fidelity (Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0081] Those skilled in the art can understand that Figure 1 the structure shown in

[0082] does not constitute a limitation on the myocardial hypertrophy measurement system, and may include more or fewer components than shown, or combine some components, or have different component arrangements. Figure 1 As

[0083] shown, the memory 1005, as a storage medium, may include an operating system, a network communication module, a user interface module, and a myocardial hypertrophy measurement program. Figure 1In the myocardial hypertrophy measurement system shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the myocardial hypertrophy measurement system of the present invention can be arranged in the myocardial hypertrophy measurement system, and the myocardial hypertrophy measurement system calls the myocardial hypertrophy measurement program stored in the memory 1005 through the processor 1001.

[0084] Based on the above hardware structure, an embodiment of the myocardial hypertrophy measurement device of the present invention is proposed.

[0085] Referring to Figure 2 , Figure 2 is a schematic flowchart of a processing method for a first embodiment of a myocardial hypertrophy measurement device of the present invention.

[0086] In this embodiment, the myocardial hypertrophy diagnosis device includes:

[0087] An interface, configured to:

[0088] Step S10: Receive a coronary CTA image of the heart.

[0089] It should be noted that the coronary CTA (Computed Tomography Angiography) image is obtained through CT angiography and is a very important part of CT clinical applications. Due to the poor natural contrast between coronary blood vessels and their background soft tissues, conventional CT plain scans often have difficulty in showing coronary blood vessels. When performing a CTA examination, a contrast agent needs to be introduced to change the image contrast between coronary blood vessels and background tissues, thereby highlighting the coronary blood vessels. The coronary CTA image is a 3D image with good three-dimensional information. By performing three-dimensional reconstruction on the coronary CTA, the three-dimensional spatial information of the coronary blood vessels can be presented intuitively and stereoscopically.

[0090] A processor, configured to:

[0091] Step S20: Based on the coronary CTA image, respectively obtain a myocardial segmentation binary map of the heart and a heart segmentation map of the heart.

[0092] In one embodiment, the processor is further configured to:

[0093] Input the coronary CTA image into a trained first deep convolutional network for myocardial segmentation to obtain a myocardial segmentation binary map of the heart;

[0094] Input the coronary CTA image into a trained second deep convolutional network for heart segmentation to obtain a heart segmentation map of the heart.

[0095] It should be noted that a large number of coronary CTA images can be input into the first deep convolutional network for training to obtain a trained first deep convolutional network. Also, a large number of coronary CTA images can be input into the second deep convolutional network for training to obtain a trained second deep convolutional network.

[0096] Step S30: Based on the heart segmentation map, determine the binary map of the left atrioventricular junction position and the binary map of the right atrioventricular junction position.

[0097] It should be noted that the atrioventricular junction position refers to the junction position between the atrium and the ventricle of the heart.

[0098] It should be noted that the left atrioventricular junction position refers to the junction position between the left atrium and the left ventricle, and the right atrioventricular junction position refers to the junction position between the right atrium and the right ventricle.

[0099] In one embodiment, the heart segmentation map includes the left ventricle, the right ventricle, the left atrium, and the right atrium; wherein,

[0100] The processor is further configured to:

[0101] Dilate the left ventricle and the left atrium respectively. After the intersection of the dilated left ventricle and the dilated left atrium is not an empty set, determine the binary map of the left atrioventricular junction position based on the intersection;

[0102] Dilate the right ventricle and the right atrium respectively. After the intersection of the dilated right ventricle and the dilated right atrium is not an empty set, determine the binary map of the right atrioventricular junction position based on the intersection.

[0103] It should be noted that dilation refers to an image processing operation method to expand the boundary of a binary object, merge all background points in contact with the object into the object, and expand the boundary of the object outward.

[0104] In a specific implementation, as Figure 3 shown, Figure 3 the left figure in it refers to the heart segmentation map, and the right figure refers to the binary map of the left atrioventricular junction position and the binary map of the right atrioventricular junction position.

[0105] In a specific implementation, the dilation rate can be set in advance, and then the left ventricle and the left atrium are dilated respectively according to the dilation rate. When the intersection of the dilated left ventricle and the dilated left atrium is an empty set, it is necessary to continue to dilate the dilated left ventricle and the dilated left atrium until the intersection of the dilated left ventricle and the dilated left atrium is not an empty set, and then determine the binary map of the left atrioventricular junction position according to the intersection of the dilated left ventricle and the dilated left atrium.

[0106] In a specific implementation, the dilation rate can be preset in advance, and then the right ventricle and the right atrium are dilated respectively according to the dilation rate. When the intersection of the dilated right ventricle and the dilated right atrium is an empty set, it is necessary to continue to dilate the dilated right ventricle and the dilated right atrium until the intersection of the dilated right ventricle and the dilated right atrium is not an empty set. Then, the binary image of the right atrioventricular junction position is determined according to the intersection of the dilated right ventricle and the dilated right atrium.

[0107] In this embodiment, by dilating the left atrium and the left ventricle in the cardiac segmentation image simultaneously to obtain the intersection of the dilated left ventricle and the dilated left atrium, and then determining the centroid of the left atrioventricular junction position based on the intersection, the centroid of the left atrioventricular junction position can be determined quickly and accurately.

[0108] Step S40: Determine the centroid of the left atrioventricular junction position according to the binary image of the left atrioventricular junction position, and determine the centroid of the right atrioventricular junction position according to the binary image of the right atrioventricular junction position.

[0109] Step S50: Determine the point on the myocardium with the maximum distance from the centroid of the left atrioventricular junction position, and determine the center point between the centroid of the left atrioventricular junction position and the maximum distance point, where the maximum distance point is determined as the apex position.

[0110] It should be noted that the distance between the center point and the centroid of the left atrioventricular junction position is 0.5O 左房室 O 心尖 。

[0111] It can be understood that the apex position can also be obtained by other means. Exemplarily, methods such as position detection can be used.

[0112] Step S60: Obtain the short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial annular segmentation map, where the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point.

[0113] It should be noted that since the short-axis plane is a plane perpendicular to the line connecting the first reference point and the second reference point and passing through the center point, the first reference point can be the centroid of the left atrioventricular junction position, and the second reference point can be the apex position. No matter which one is used as the first reference point and the second reference point to obtain the short-axis plane, it is used to set the standard for judging the myocardial thickness, and the choice of the reference point position is not unique.

[0114] It should be noted that the short-axis plane can also be determined based on other reference points inspired by the centroid of the left atrioventricular junction position and the apex position. Preferably, such as the centroids of the left and right atria, etc.

[0115] Step S70: Determine the long-axis plane based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the maximum distance point.

[0116] Step S80: Intercept the short-axis myocardial annular segmentation map based on the long-axis plane to obtain a target connected component.

[0117] It should be noted that intercepting the short-axis myocardial annular segmentation map based on the long-axis plane can obtain two connected components, and the target connected component can be determined from the two connected components.

[0118] In a specific implementation, as Figure 4 shown, the long-axis plane is used to intercept the short-axis myocardial annular segmentation map to obtain two connected components.

[0119] In one embodiment, the processor is further configured to:

[0120] Determine the centroid of the right ventricle according to the cardiac segmentation map;

[0121] Intercept the short-axis myocardial annular segmentation map based on the long-axis plane to obtain a first connected component and a second connected component;

[0122] Determine a first distance from the centroid of the right ventricle to the midpoint of the first connected component, and determine a second distance from the centroid of the right ventricle to the midpoint of the second connected component;

[0123] Determine the smaller distance between the first distance and the second distance, and use the connected component corresponding to the smaller distance as the target connected component.

[0124] It should be noted that the target connected component can be understood as the connected component closest to the centroid of the right ventricle or the centroid of the right atrium.

[0125] Step S90: Determine the positions of the two endpoints in the target connected component, and determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the measurement value of the myocardial thickness at the middle part.

[0126] In one embodiment, the processor is further configured to:

[0127] Based on the straight-line feature of a single pixel in the target connected component, determine the number of pixels occupied by the maximum distance between any two points in the target connected component;

[0128] Determine the myocardial thickness measurement value according to the occupied number of pixels.

[0129] In a specific implementation, the myocardial thickness measurement value can be determined according to the product of the distance occupied by each pixel and the number of pixels.

[0130] In one embodiment, the processor is further configured to:

[0131] Determine the diagnostic threshold for myocardial hypertrophy;

[0132] Determine whether the measured value of the myocardial thickness at the middle is greater than the diagnostic threshold for myocardial hypertrophy;

[0133] If it is determined that the measured value of the myocardial thickness at the middle is greater than the diagnostic threshold for myocardial hypertrophy, then diagnose the patient corresponding to the heart as a patient with myocardial hypertrophy.

[0134] It should be noted that clinically, the diagnosis of myocardial hypertrophy usually takes the myocardial thickness at the position near the interventricular groove as the standard. When the myocardial thickness at this position is greater than 15 mm, it can be diagnosed as myocardial hypertrophy. Preferably, the diagnostic threshold for myocardial hypertrophy can be set to 15 mm.

[0135] In this embodiment, by receiving the coronary CTA image of the heart; based on the coronary CTA image, respectively obtaining the myocardial segmentation binary map of the heart and the heart segmentation map of the heart; based on the heart segmentation map, determining the left atrioventricular junction position binary map and the right atrioventricular junction position binary map; determining the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determining the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map; determining the point with the maximum distance from the myocardium to the centroid of the left atrioventricular junction position, and determining the center point between the centroid of the left atrioventricular junction position and the point with the maximum distance, wherein the point with the maximum distance is determined as the apex position; obtaining the short-axis plane, and intercepting the myocardium based on the short-axis plane to obtain a short-axis myocardial circular segmentation map, wherein the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point; based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the apex position, determining the long-axis plane; based on the long-axis plane, intercepting the short-axis myocardial circular segmentation map to obtain a target connected region; determining the positions of the two end points in the target connected region, and determining the distance between the two end points according to the positions of the two end points, and taking the distance as the measured value of the myocardial thickness at the middle. Through the above method, the myocardial thickness can be measured quickly and accurately.

[0136] Reference Figure 5 , Figure 5 is a schematic diagram of the processing method flow of the second embodiment of a myocardial hypertrophy measurement device according to the present invention.

[0137] Based on the above first embodiment, the processing method of this embodiment further includes:

[0138] Step S110: Determine the line segment O between the centroid of the left atrioventricular junction position and the point with the maximum distance 左房室 O 心尖the first target point therein, where the distance between the first target point and the centroid of the left atrioventricular junction position is between 0.2 - 0.3O 左房室 O 心尖 between.

[0139] It should be noted that the myocardial thickness corresponding to any point can be obtained according to the needs of the doctor. Specifically, the corresponding myocardial thickness can be obtained by adjusting the position of the first target point. Preferably, when the distance between the first target point and the centroid of the left atrioventricular junction position is 0.25O 左房室 O 心尖 , the myocardial thickness measurement value at the heart base can be accurately determined.

[0140] Step S111: Obtain the first short-axis plane, and based on the first short-axis plane, intercept the myocardium to obtain a first short-axis myocardial annular segmentation map, where the first short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the first target point.

[0141] Step S112: Based on the long-axis plane, intercept the first short-axis myocardial annular segmentation map to obtain a target connected region.

[0142] It should be noted that intercepting the first short-axis myocardial annular segmentation map based on the long-axis plane can obtain two connected regions, and the target connected region can be determined from the two connected regions.

[0143] It should be noted that the target connected region can be understood as the connected region closest to the centroid of the right ventricle or the centroid of the right atrium.

[0144] Step S113: Determine the positions of the two endpoints in the target connected region, and determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the myocardial thickness measurement value at the heart base.

[0145] In a specific implementation, the myocardial thickness measurement value at the heart base can be determined by calculating the spatial distance between the coordinates of the two endpoints of the target connected region.

[0146] In one embodiment, the processor is further configured to:

[0147] Based on the straight-line characteristics of a single pixel in the target connected region, determine the number of pixels occupied by the maximum distance between any two points in the target connected region;

[0148] According to the number of occupied pixels, determine the myocardial thickness measurement value at the heart base.

[0149] In a specific implementation, the myocardial thickness measurement value at the heart base can be determined according to the product of the distance occupied by each pixel and the number of pixels.

[0150] In this embodiment, the line segment O between the centroid of the left atrioventricular junction position and the maximum distance point is determined 左房室 O 心尖 and the first target point in it, where the distance between the first target point and the centroid of the left atrioventricular junction position is between 0.2 - 0.3O 左房室 O 心尖 ; the first short-axis plane is obtained, and the first short-axis myocardial annular segmentation map is obtained by intercepting the myocardium based on the first short-axis plane, where the first short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the first target point; based on the long-axis plane, the first short-axis myocardial annular segmentation map is intercepted to obtain the target connected region; the maximum distance between any two points in the target connected region is determined as the myocardial thickness measurement value at the cardiac base. In the above manner, the myocardial thickness measurement value at the cardiac base can be determined quickly and accurately.

[0151] Reference Figure 6 , Figure 6 is a schematic flowchart of the processing method of the third embodiment of a myocardial hypertrophy measurement device of the present invention.

[0152] Based on the above first embodiment, the processing method of this embodiment further includes:

[0153] Step S120: Determine the second target point on the line segment O between the centroid of the left atrioventricular junction position and the maximum distance point 左房室 O 心尖 where the distance between the second target point and the centroid of the left atrioventricular junction position is between 0.7 - 0.8O 左房室 O 心尖 between.

[0154] It should be noted that the myocardial thickness corresponding to any point can be obtained according to the doctor's needs. Specifically, the corresponding myocardial thickness can be obtained by adjusting the position of the second target point. Preferably, when the distance between the first target point and the centroid of the left atrioventricular junction position is 0.75O 左房室 O 心尖 the myocardial thickness measurement value at the cardiac apex can be accurately determined.

[0155] Step S121: Obtain the second short-axis plane, and intercept the myocardium based on the second short-axis plane to obtain the second short-axis myocardial annular segmentation map, where the second short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the second target point.

[0156] Step S122: Based on the long-axis plane, intercept the second short-axis myocardial annular segmentation map to obtain the target connected region.

[0157] It should be noted that intercepting the first short-axis myocardial annular segmentation map based on the long-axis plane can obtain two connected domains, and the target connected domain can be determined from the two connected domains.

[0158] It should be noted that the target connected domain can be understood as the connected domain closest to the centroid of the right ventricle or the centroid of the right atrium.

[0159] Step S123: Perform skeletonization processing on the target connected domain to obtain the skeletonized target connected domain.

[0160] Step S124: Determine the positions of two endpoints based on the skeletonized target connected domain, determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the measurement value of the myocardial thickness at the apex.

[0161] In one embodiment, the processor is further configured to:

[0162] Based on the straight-line characteristics of a single pixel in the target connected domain, determine the number of pixels occupied by the maximum distance between any two points in the target connected domain;

[0163] Determine the measurement value of the myocardial thickness at the apex according to the occupied number of pixels.

[0164] In a specific implementation, the measurement value of the myocardial thickness at the apex can be determined according to the product of the distance occupied by each pixel and the number of pixels.

[0165] In this embodiment, by determining the second target point on the line segment O 左房室 O 心尖 between the centroid of the left atrioventricular junction position and the maximum distance point, where the distance between the second target point and the centroid of the left atrioventricular junction position is between 0.7 - 0.8O 左房室 O 心尖 ; obtaining the second short-axis plane, and intercepting the myocardium based on the second short-axis plane to obtain the second short-axis myocardial annular segmentation map, where the second short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the second target point; based on the long-axis plane, intercept the second short-axis myocardial annular segmentation map to obtain the target connected domain; perform skeletonization processing on the target connected domain to obtain the skeletonized target connected domain; determine the positions of two endpoints based on the skeletonized target connected domain, determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the measurement value of the myocardial thickness at the apex. Through the above method, the measurement value of the myocardial thickness at the apex can be measured quickly and accurately.

[0166] It should be noted that the workflow described above is only illustrative and does not limit the protection scope of the present invention. In actual applications, those skilled in the art can select some or all of them according to actual needs to achieve the purpose of the solution of this embodiment, and no limitation is made here.

[0167] In addition, it should be noted that in this text, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or system including that element.

[0168] The serial numbers of the above-described embodiments of the present invention are only for description and do not represent the superiority or inferiority of the embodiments.

[0169] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as a read-only memory (ROM) / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.

[0170] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A myocardial hypertrophy measurement device, characterized in that, The myocardial hypertrophy diagnosis device includes: An interface configured to receive coronary CTA images of the heart; A processor configured to: Based on the coronary CTA images, respectively obtain a myocardial segmentation binary map of the heart and a heart segmentation map of the heart; Based on the heart segmentation map, determine a left atrioventricular junction position binary map and a right atrioventricular junction position binary map; Determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map; Determine the point on the myocardium with the maximum distance to the centroid of the left atrioventricular junction position, and determine the center point between the centroid of the left atrioventricular junction position and the maximum distance point, wherein the maximum distance point is determined as the apex position; Obtain a short-axis plane, and based on the short-axis plane, intercept the myocardium to obtain a short-axis myocardial circular segmentation map, wherein the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point; Based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the apex position, determine a long-axis plane; Based on the long-axis plane, intercept the short-axis myocardial circular segmentation map to obtain a target connected component; Determine the positions of two endpoints in the target connected component, and determine the distance between the two endpoints according to the positions of the two endpoints, and use the distance as the myocardial thickness measurement value at the middle part.

2. The device according to claim 1, characterized in that, The heart segmentation map includes the left ventricle, the right ventricle, the left atrium, and the right atrium; wherein, The processor is further configured to: Respectively dilate the left ventricle and the left atrium, and after making the intersection of the dilated left ventricle and the dilated left atrium non-empty, determine the left atrioventricular junction position binary map based on the intersection; Respectively dilate the right ventricle and the right atrium, and after making the intersection of the dilated right ventricle and the dilated right atrium non-empty, determine the right atrioventricular junction position binary map based on the intersection.

3. The device according to claim 2, wherein The processor is further configured to: Determine the centroid of the right ventricle according to the heart segmentation map; Based on the long-axis plane, intercept the short-axis myocardial circular segmentation map to obtain a first connected component and a second connected component; Determine a first distance from the centroid of the right ventricle to the midpoint of the first connected component, and determine a second distance from the centroid of the right ventricle to the midpoint of the second connected component; Determine the smaller distance between the first distance and the second distance, and use the connected component corresponding to the smaller distance as the target connected component.

4. The device according to claim 1, wherein The processor is further configured to: Based on the linear characteristics of a single pixel in the target connected component, determine the number of pixels occupied by the maximum distance between any two points in the target connected component; Determine the myocardial thickness measurement value according to the occupied number of pixels.

5. The device according to claim 1, characterized in that, The processor is further configured to: Input the coronary CTA image into a trained first deep convolutional network to segment the myocardium, and obtain a myocardial segmentation binary map of the heart; Input the coronary CTA image into a trained second deep convolutional network to segment the heart, and obtain a heart segmentation map of the heart.

6. The device according to claim 1, characterized in that, The processor is further configured to: Determine the diagnostic threshold for myocardial hypertrophy; Judge whether the measured value of the myocardial thickness at the middle is greater than the diagnostic threshold for myocardial hypertrophy; If it is determined that the measured value of the myocardial thickness at the middle is greater than the diagnostic threshold for myocardial hypertrophy, diagnose the patient corresponding to the heart as a patient with myocardial hypertrophy.

7. The device according to claim 1, characterized in that The processor is further configured to: Determine the line segment O between the centroid of the left atrioventricular junction position and the maximum distance point 左房室 O 心尖 The first target point in it, where the distance between the first target point and the centroid of the left atrioventricular junction position is between 0.2 - 0.3O 左房室 O 心尖 ; Obtain a first short-axis plane, and intercept the myocardium based on the first short-axis plane to obtain a first short-axis myocardial circular segmentation map, wherein the first short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the first target point; Based on the long-axis plane, intercept the first short-axis myocardial circular segmentation map to obtain a target connected region; Determine the positions of two end points in the target connected region, and determine the distance between the two end points according to the positions of the two end points, and use the distance as the measured value of the myocardial thickness at the base of the heart.

8. The device according to claim 1, characterized in that, The processor is further configured to: Determine the line segment O between the centroid of the left atrioventricular junction position and the maximum distance point 左房室 O 心尖 The second target point in it, where the distance between the second target point and the centroid of the left atrioventricular junction position is between 0.7 - 0.8O 左房室 O 心尖 ; Obtain a second short-axis plane, and intercept the myocardium based on the second short-axis plane to obtain a second short-axis myocardial circular segmentation map, wherein the second short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the maximum distance point and passing through the second target point; Based on the long-axis plane, intercept the second short-axis myocardial circular segmentation map to obtain a target connected region; Perform skeletonization processing on the target connected region to obtain a skeletonized target connected region; Based on the skeletonized target connected region, determine the positions of two end points, determine the distance between the two end points according to the positions of the two end points, and use the distance as the measured value of the myocardial thickness at the apex of the heart.

9. A myocardial hypertrophy measurement system, characterized in that, The system includes: The measuring device according to any one of claims 1 to 8; The myocardial hypertrophy diagnosis device includes: An interface configured to receive a coronary CTA image of the heart; A processor configured to: Based on the coronary CTA image, respectively obtain a myocardial segmentation binary map of the heart and a heart segmentation map of the heart; Based on the heart segmentation map, determine a left atrioventricular junction position binary map and a right atrioventricular junction position binary map; Determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map; Determine the maximum distance point on the myocardium to the centroid of the left atrioventricular junction position, and determine the center point of the centroid of the left atrioventricular junction position and the maximum distance point, wherein the maximum distance point is determined as the apex position; Obtain a short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial circular segmentation map, wherein the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point; Based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the apex position, determine a long-axis plane; Based on the long-axis plane, intercept the short-axis myocardial circular segmentation map to obtain a target connected region; Determine the positions of two end points in the target connected region, and determine the distance between the two end points according to the positions of the two end points, and use the distance as the myocardial thickness measurement value at the middle part.

10. A storage medium, characterized in that, A myocardial hypertrophy measurement program is stored on the storage medium. When the myocardial hypertrophy measurement program is executed by a processor, the following steps are implemented: Based on the coronary CTA image, respectively obtain the myocardial segmentation binary map of the heart and the heart segmentation map of the heart; Based on the heart segmentation map, determine the left atrioventricular junction position binary map and the right atrioventricular junction position binary map; Determine the centroid of the left atrioventricular junction position according to the left atrioventricular junction position binary map, and determine the centroid of the right atrioventricular junction position according to the right atrioventricular junction position binary map; Determine the point on the myocardium with the maximum distance from the centroid of the left atrioventricular junction position, and determine the center point between the centroid of the left atrioventricular junction position and the maximum distance point, where the maximum distance point is determined as the apex position; Obtain the short-axis plane, and intercept the myocardium based on the short-axis plane to obtain a short-axis myocardial annular segmentation map, where the short-axis plane is a plane perpendicular to the line connecting the centroid of the left atrioventricular junction position and the apex position and passing through the center point; Based on the centroid of the left atrioventricular junction position, the centroid of the right atrioventricular junction position, and the apex position, determine the long-axis plane; Based on the long-axis plane, intercept the short-axis myocardial annular segmentation map to obtain a target connected region; Determine the positions of two end points in the target connected region, and determine the distance between the two end points according to the positions of the two end points, and use the distance as the myocardial thickness measurement value at the middle part.

Images