A left ventricular segment myocardial CT value acquisition method and system

Abstract

The application discloses a left ventricular segment myocardial CT value acquisition method and system, relates to the technical field of myocardial CT value measurement, and comprises the following steps: acquiring complete left ventricular image data; using different algorithms to slice reconstruction of the left ventricular image data, so as to obtain a cardiac CT delay enhancement image of each patient; preprocessing the cardiac CT delay enhancement image; using a 3DnnUNet model to segment the preprocessed cardiac CT delay enhancement image, so as to obtain a segmentation result of a left ventricular myocardial region; further dividing the segmented left ventricular myocardial region; acquiring a pixel mean value of each region, and then converting the pixel mean value into a corresponding CT mean value. The left ventricular myocardial region in the cardiac CT delay enhancement image is accurately segmented, the segmentation result is regionally divided according to an AHA 16-segment model, and the CT value of each region is obtained; and the application can provide more objective and accurate myocardial CT value measurement results, and provide reliable support for clinical diagnosis and treatment.

Description

Technical Field

[0001] This invention relates to the field of myocardial CT value measurement technology, specifically to a method and system for obtaining segmental myocardial CT values ​​of the left ventricle. Background Technology

[0002] With the continued rise in the incidence of cardiovascular diseases, early diagnosis and precise treatment of heart disease have become particularly important. Cardiac CT (computed tomography) imaging technology, with its high resolution and rapid scanning capabilities, has become a key tool for assessing cardiac structure and function. In particular, delayed-contrast cardiac CT (CT-LE) imaging technology, through the injection of contrast agents, can clearly show the extent of myocardial fibrosis and myocardial infarction, thus providing a clearer picture of the degree of myocardial damage.

[0003] The 17-segment and modified 16-segment models proposed by the American Heart Association (AHA) are important classification methods used in clinical and research work for standardized assessment of precise localization of myocardial injury and overall and segmental cardiac function. The 16-segment model divides the left ventricle (LV) into 16 anatomical segments for detailed assessment of the function and pathological status of each segment. Specifically, the LV is vertically divided into three equal parts along the long axis of the heart, generating three circular portions of the LV: the basal level, the papillary muscle level, and the apical level. Only myocardial slices encompassing all 360° of the left ventricle are considered. The basal image is further divided into six 60° segments, named: basal anterior wall, basal anterior septum, basal inferior septum, basal inferior wall, basal inferior lateral wall, and basal anterior lateral wall. Similarly, the papillary muscle image is also divided into six 60° segments, called the intermediate anterior wall, intermediate anterior septum, intermediate inferior septum, intermediate inferior wall, intermediate inferior lateral wall, and intermediate anterior lateral wall. Because the left ventricle is cone-shaped, the apical image uses only four 90° segments, named the anterior apical wall, apical septum, inferior apical wall, and lateral apical wall. The apical operculum represents the actual muscle at the extreme tip of the ventricle, where there are no longer any chambers; this part is called the apex.

[0004] Accurate assessment of the location and extent of left ventricular myocardial injury is crucial for the diagnosis and treatment of cardiovascular diseases. However, current technologies still have the following shortcomings: First, the accuracy and consistency of manual segmentation methods are insufficient. Traditional myocardial region segmentation mainly relies on manual delineation. However, due to the low contrast between the blood pool and myocardium in delayed-enhanced cardiac CT images, and the presence of interfering factors such as trabeculae and papillary muscles in the endocardium, manual segmentation is time-consuming and prone to subjective errors, resulting in low accuracy and consistency of the segmentation results. Second, the segmentation precision is not high. Existing segmentation methods struggle to accurately divide different layers of myocardial regions, such as the apex, papillary muscles, and basal region, leading to coarse and inconsistent segmentation results. This makes it impossible to provide stable and reliable CT value calculation results, which may affect clinicians' accurate assessment of the condition and prognosis. Finally, CT value calculation is inaccurate. In existing technologies, the conversion between pixel values ​​and CT values ​​often relies on automatic processing by the equipment. However, due to the poor signal-to-noise ratio and image quality of delayed-enhanced cardiac CT images, if the equipment fails to convert accurately, manual calculation is required, further increasing workload and the risk of error.

[0005] In recent years, deep learning technology, especially convolutional neural networks (CNNs), has made significant progress in the field of medical image segmentation. Among them, U-Net and its variants have become the mainstream methods for medical image segmentation. 3D nnU-Net, as a three-dimensional network architecture based on U-Net, further improves the accuracy and robustness of medical image segmentation. Summary of the Invention

[0006] The purpose of this invention is to propose a method and system for obtaining segmental myocardial CT values ​​of the left ventricle, which can provide more objective and accurate myocardial CT value measurement results, providing reliable support for clinical diagnosis and treatment.

[0007] According to a first aspect of the present disclosure, a method for obtaining segmental myocardial CT values ​​of the left ventricle is provided, comprising the following steps:

[0008] Delayed-enhanced CT scans of the left ventricle were performed on multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient;

[0009] The cardiac CT delayed enhancement images are preprocessed, including image normalization and noise removal, to improve image quality;

[0010] The 3DnnUNet model was used to segment preprocessed delayed-enhanced cardiac CT images to obtain the segmentation results of the left ventricular myocardial region. 3DnnUNet is a deep learning-based segmentation model that can accurately segment images in three-dimensional space.

[0011] The segmented left ventricular myocardial region was further subdivided: the apical CT delayed enhancement image was divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image was divided into 6 regions: the anterior wall of the middle part, the anterior septum of the middle part, the inferior septum of the middle part, the inferior wall of the middle part, the inferior lateral wall of the middle part, and the anterior lateral wall of the middle part; the basal CT delayed enhancement image was divided into 6 regions: the anterior wall of the base, the anterior septum of the base, the inferior septum of the base, the inferior wall of the base, the inferior lateral wall of the base, and the anterior lateral wall of the basal segment.

[0012] The pixel mean of each region is obtained and then converted into the corresponding CT mean.

[0013] According to a second aspect of the present disclosure, a system for acquiring segmental myocardial CT values ​​of the left ventricle is provided, comprising:

[0014] A dataset module was constructed to perform delayed-enhanced CT scans of the left ventricle of multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient;

[0015] The image preprocessing module preprocesses the delayed-enhancement cardiac CT image, including image normalization and noise removal, to improve image quality.

[0016] The myocardial region segmentation module uses the 3DnnUNet model to segment preprocessed delayed-enhanced cardiac CT images to obtain the segmentation results of the left ventricular myocardial region. 3DnnUNet is a deep learning-based segmentation model that can accurately segment images in three-dimensional space.

[0017] The region segmentation module further subdivides the segmented left ventricular myocardial region: the apical CT delayed enhancement image is divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image is divided into 6 regions: the anterior wall of the middle part, the anterior septum of the middle part, the inferior septum of the middle part, the inferior wall of the middle part, the inferior lateral wall of the middle part, and the anterior lateral wall of the middle part; the basal CT delayed enhancement image is divided into 6 regions: the anterior wall of the basement, the anterior septum of the basement, the inferior septum of the basement, the inferior wall of the basement, the inferior lateral wall of the basement, and the anterior lateral wall of the basal segment.

[0018] The CT value acquisition module calculates the CT value for each region: it obtains the pixel mean for each region and then converts it into the corresponding CT mean.

[0019] According to a third aspect of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the memory, wherein the processor executes the program to implement the method for obtaining left ventricular segmental myocardial CT values.

[0020] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the method for obtaining CT values ​​of left ventricular segmental myocardium.

[0021] Compared with the prior art, the above technical solutions adopted in this invention have the following advantages: 1. Improved accuracy and consistency of segmentation: Automated segmentation using the 3DnnUNet model greatly improves the accuracy and consistency of myocardial region segmentation and reduces errors caused by human operation.

[0022] 2. Detailed regional division and CT value calculation: Based on the 16-segment model proposed by the American Heart Association (AHA), the myocardial region is divided in detail, providing more accurate CT values ​​to quantify myocardial damage, which helps clinicians to make accurate diagnoses and treatments.

[0023] 3. Saves time and improves efficiency: Compared with traditional manual segmentation methods, this invention significantly shortens the time for image processing and data analysis, and improves work efficiency.

[0024] 4. Automated CT value conversion and calculation: Automatically reads DICOM information and performs CT value conversion, ensuring the accuracy of CT value calculation, reducing the risk of errors from manual calculation, and further improving the accuracy and efficiency of diagnosis. Attached Figure Description

[0025] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an undue limitation of this application.

[0026] Figure 1 Flowchart of the method for obtaining segmental myocardial CT values ​​of the left ventricle;

[0027] Figure 2 Here is a diagram of the 3DnnUNet model structure;

[0028] Figure 3 This is a flowchart illustrating the method for obtaining CT values. Specific implementation methods

[0029] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0030] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0031] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0032] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of this disclosure. It should be noted that each block in a flowchart or block diagram may represent a module, segment, or portion of code, which may include one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutively represented blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0033] Example 1:

[0034] This embodiment provides a method for obtaining segmental myocardial CT values ​​of the left ventricle, including the following steps:

[0035] S1: Perform delayed-enhanced CT scans on the left ventricle of multiple patients to obtain complete left ventricular image data; use different algorithms to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient;

[0036] Specifically, doctors select multiple patients to ensure the representativeness and diversity of the sample. Then, delayed-contrast CT scans are performed on the left ventricle of each patient to obtain complete left ventricular imaging data. Different algorithms can be used, such as the Aice (Advanced Intelligent Clear-IQ Engine) algorithm based on deep learning reconstruction, the MBIR (Model-Based Iterative Reconstruction) algorithm for model iterative reconstruction, and the HIR (Hybrid Iterative Reconstruction) algorithm for hybrid iterative reconstruction. These algorithms reconstruct slices of the left ventricle, each approximately 10mm thick, covering the entire left ventricle from top to bottom. These three different algorithms ensure the diversity and comprehensiveness of the data. Finally, delayed-contrast CT images of the heart are obtained for each patient, providing the foundational data for subsequent image processing and analysis.

[0037] S2: Preprocess the cardiac CT delayed enhancement image, including image normalization and noise removal, to improve image quality;

[0038] Specifically, a Gaussian filter is used to smooth the delayed-enhanced cardiac CT images; the Gaussian filter can effectively reduce high-frequency noise while preserving the edge information of the image.

[0039] I'=G*I

[0040] Where I is the original image, G is the Gaussian kernel, I' is the filtered image, and * indicates the convolution operation. The parameters of the Gaussian kernel (such as the standard deviation σ) are adjusted to achieve optimal noise removal while avoiding over-smoothing that could lead to loss of detail.

[0041] S3: The 3DnnUNet model is used to segment the preprocessed delayed-enhancement cardiac CT images to obtain the segmentation results of the left ventricular myocardial region; specifically, this includes:

[0042] S31: Classify the preprocessed cardiac CT delayed enhancement images into apical, papillary muscle, and basal categories;

[0043] S32: Training the 3DnnUNet model for the apex category: Using the first 3DnnUNet architecture, network parameters such as kernel size, number of layers, and learning rate are set; the model is trained using the cross-entropy loss function and the Adam optimizer. Data augmentation (such as rotation and translation) is performed during training to increase the model's generalization ability. The training data consists of delayed-enhanced cardiac CT images of the apex category and their corresponding labels; after training, the optimal model weights are saved, and the delayed-enhanced cardiac CT images of the apex category are input into the optimal 3DnnUNet model to obtain segmented images of the apex myocardial region;

[0044] S33: Training 3DnnUNet models for papillary muscle and basal muscle categories: Using the second 3DnnUNet architecture, network parameters such as kernel size, number of layers, and learning rate are set; the model is trained using the cross-entropy loss function and the Adam optimizer. Data augmentation (such as rotation and translation) is performed during training to increase the model's generalization ability. The training data consists of delayed-enhanced cardiac CT images of the papillary muscle category, delayed-enhanced cardiac CT images of the basal muscle category, and their corresponding labels; after training, the optimal model weights are saved, and the delayed-enhanced cardiac CT images of the papillary muscle and basal muscle categories are input into the optimal 3DnnUNet model to obtain segmented images of the papillary muscle myocardial region and the basal myocardial region;

[0045] Because the apex class differs significantly from the papillary muscle and basal class in terms of anatomical structure and image features, two independent 3DnnUNet models were trained separately.

[0046] S34: The segmentation results of the left ventricular myocardial region are obtained by stitching together the original DICOM images of the apex, papillary muscles, and base of the heart with the corresponding myocardial region segmentation labels of these three images.

[0047] S4: Further subdivide the segmented left ventricular myocardial region: Divide the apical CT delayed enhancement image into 4 regions: apical anterior wall, apical septum, apical inferior wall, and apical lateral wall; divide the papillary muscle CT delayed enhancement image into 6 regions: middle anterior wall, middle anterior septum, middle inferior septum, middle inferior wall, middle inferior lateral wall, and middle anterior lateral wall; divide the basal CT delayed enhancement image into 6 regions: basal anterior wall, basal anterior septum, basal inferior septum, basal inferior wall, basal inferior lateral wall, and basal anterior lateral wall.

[0048] It should be noted that after classification, three DICOM images and segmentation mask images are selected for each patient. These three images are those considered by the physician to be the most representative of the apex, papillary muscle, and basal segment categories. The specific steps are as follows:

[0049] S41: Use the pydicom.dcmread function to read three DICOM images; use the nib.load function to read the segmentation mask image label (NIFTI format) corresponding to each image; the three DICOM images are the most representative segmentation images of the apical myocardium, papillary myocardium, and basal myocardium regions;

[0050] S42: Adjust the labels of the read DICOM image and segmentation mask image so that their positions correspond to each other;

[0051] S43: Traverse each segmentation mask image label and record the positions where the pixel value is 0; also set the pixel value of the corresponding pixel in the DICOM image to 0, thereby determining the region on the DICOM image where the CT value needs to be calculated;

[0052] S44: The trim.py function is used to crop out the excess black background in an image and reshape the cropped image into a square.

[0053] S45: The functions get6.py and get4.py are used to obtain two regions. In the get6.py function, the delayed enhancement CT image of the basal and papillary muscles is divided into 6 parts according to the 16-segment model proposed by the American Heart Association (AHA). It first defines the vertex coordinates of the 6 sector regions, and then divides the input image into 6 parts and stores them in the segment list. Similarly, in the get4.py function, the delayed enhancement CT image of the apex is divided into 4 parts according to the 16-segment model proposed by the American Heart Association (AHA). It first defines the vertex coordinates of the 4 sector regions, and then divides the input image into 4 parts and stores them in the segment list.

[0054] S5: Obtain the pixel mean of each region and then convert it into the corresponding CT mean.

[0055] Medical CT images differ from natural images; their pixel matrix represents reconstructed data of CT values ​​or tissue relaxation times. For example, in CT, the CT value of water is 0, and its calculation formula is:

[0056]

[0057] In the formula, μ and μ w These are the attenuation coefficients for the tested object and water, respectively. Some devices do not convert pixel values ​​to CT values ​​after scanning, so manual conversion is required. The conversion formula between pixel mean and CT mean is:

[0058] CT = pixel × slope + intercept

[0059] Where pixel is the number of pixels in the delayed-enhanced cardiac CT image, slope is the slope of the delayed-enhanced cardiac CT image, and intercept is the intercept.

[0060] To obtain the CT value of an image, two DICOM label information items need to be read first: the rescale intercept and the rescale slope. This conversion is very useful in practice because the severity of certain diseases is correlated with the magnitude of the CT value in CT images. After segmentation or localization using machine learning, the magnitude of the CT value often becomes one of the important output indicators. The specific steps are as follows:

[0061] S51: Obtain the rescale intercept and rescale slope for each DICOM image;

[0062] S52: Iterate through the 6 regions of the basal CT delayed-enhanced image and the 6 regions of the papillary muscle CT delayed-enhanced image output by the get6.py function, read the pixel value of each pixel in each region, and then obtain the pixel mean. Use the conversion formula to convert the pixel mean of each region to the CT mean. Iterate through the 4 regions of the apex CT delayed-enhanced image output by the get4.py function, read the pixel value of each pixel in each region, and then obtain the pixel mean. Use the conversion formula to convert the pixel mean of each region to the CT mean.

[0063] S53: Store the average CT values ​​of 6 regions in the delayed enhancement image of the basal CT, 6 regions in the papillary muscle category, and 4 regions in the delayed enhancement image of the apex CT in the list averagesCT;

[0064] S54: Map the mean CT values ​​to 16 regions and draw a 16-segment model;

[0065] S55: The stack.py function is used to stack the segmented mask images of each myocardial region together, making it easier for doctors to observe.

[0066] Example 2:

[0067] This embodiment provides a system for acquiring segmental myocardial CT values ​​of the left ventricle, including:

[0068] A dataset module was constructed to perform delayed-enhanced CT scans of the left ventricle of multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient;

[0069] The image preprocessing module preprocesses the delayed-enhancement cardiac CT image, including image normalization and noise removal, to improve image quality.

[0070] The myocardial region segmentation module uses the 3DnnUNet model to segment the preprocessed cardiac CT delayed enhancement image to obtain the segmentation result of the left ventricular myocardial region.

[0071] The region segmentation module further subdivides the segmented left ventricular myocardial region: the apical CT delayed enhancement image is divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image is divided into 6 regions: the anterior wall of the middle segment, the anterior septum of the middle segment, the inferior septum of the middle segment, the inferior wall of the middle segment, the inferior lateral wall of the middle segment, and the anterior lateral wall of the middle segment; the myocardial portion of the basal slice is divided into 6 regions: the anterior wall of the base, the anterior septum of the base, the inferior septum of the base, the inferior wall of the base, the inferior lateral wall of the base, and the anterior lateral wall of the basal segment.

[0072] The CT value acquisition module calculates the CT value for each region: it obtains the pixel mean for each region and then converts it into the corresponding CT mean.

[0073] Example 3:

[0074] An electronic device includes a memory, a processor, and a computer program stored in the memory and running thereon. When the processor executes the program, it implements the aforementioned method for acquiring segmental myocardial CT values ​​of the left ventricle, comprising:

[0075] Delayed-enhanced CT scans of the left ventricle were performed on multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient;

[0076] The cardiac CT delayed enhancement images are preprocessed, including image normalization and noise removal;

[0077] The 3DnnUNet model was used to segment the preprocessed delayed-enhanced cardiac CT images to obtain the segmentation results of the left ventricular myocardial region;

[0078] The segmented left ventricular myocardial region was further subdivided: the apical CT delayed enhancement image was divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image was divided into 6 regions: the anterior wall of the middle part, the anterior septum of the middle part, the inferior septum of the middle part, the inferior wall of the middle part, the inferior lateral wall of the middle part, and the anterior lateral wall of the middle part; the basal CT delayed enhancement image was divided into 6 regions: the anterior wall of the base, the anterior septum of the base, the inferior septum of the base, the inferior wall of the base, the inferior lateral wall of the base, and the anterior lateral wall of the basal segment.

[0079] The pixel mean of each region is obtained and then converted into the corresponding CT mean.

[0080] Example 4:

[0081] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method for acquiring segmental myocardial CT values ​​of the left ventricle, comprising:

[0082] Delayed-enhanced CT scans of the left ventricle were performed on multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient;

[0083] The cardiac CT delayed enhancement images are preprocessed, including image normalization and noise removal;

[0084] The 3DnnUNet model was used to segment the preprocessed delayed-enhanced cardiac CT images to obtain the segmentation results of the left ventricular myocardial region;

[0085] The segmented left ventricular myocardial region was further subdivided: the apical CT delayed enhancement image was divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image was divided into 6 regions: the anterior wall of the middle part, the anterior septum of the middle part, the inferior septum of the middle part, the inferior wall of the middle part, the inferior lateral wall of the middle part, and the anterior lateral wall of the middle part; the basal CT delayed enhancement image was divided into 6 regions: the anterior wall of the base, the anterior septum of the base, the inferior septum of the base, the inferior wall of the base, the inferior lateral wall of the base, and the anterior lateral wall of the basal segment.

[0086] The pixel mean of each region is obtained and then converted into the corresponding CT mean.

[0087] Those skilled in the art will understand that the modules or steps described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, which can then be stored in a storage device for execution by a computer device. Alternatively, they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. This disclosure is not limited to any particular combination of hardware and software.

[0088] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0089] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A method for obtaining segmental CT values ​​of the left ventricular myocardium, characterized in that, Includes the following steps: Delayed-enhanced CT scans of the left ventricle were performed on multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient; The cardiac CT delayed enhancement images are preprocessed, including image normalization and noise removal; The 3DnnUNet model was used to segment the preprocessed delayed-enhanced cardiac CT images to obtain the segmentation results of the left ventricular myocardial region; The segmented left ventricular myocardial region was further subdivided: the apical CT delayed enhancement image was divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image was divided into 6 regions: the anterior wall of the middle part, the anterior septum of the middle part, the inferior septum of the middle part, the inferior wall of the middle part, the inferior lateral wall of the middle part, and the anterior lateral wall of the middle part; the basal CT delayed enhancement image was divided into 6 regions: the anterior wall of the base, the anterior septum of the base, the inferior septum of the base, the inferior wall of the base, the inferior lateral wall of the base, and the anterior lateral wall of the basal segment. Obtain the pixel mean value for each region, and then convert it into the corresponding CT mean value; The 3DnnUNet model is used to segment preprocessed delayed-enhanced cardiac CT images, specifically as follows: The preprocessed cardiac CT delayed enhancement images were divided into apex, papillary muscle, and basal categories; Training the 3DnnUNet model for the apex category: Using the first 3DnnUNet architecture, network parameters are set; the model is trained using the cross-entropy loss function and the Adam optimizer, and data augmentation is performed during training to increase the model's generalization ability; After training, the optimal model weights are saved, and the delayed-enhanced cardiac CT images of the apex category are input into the optimal 3DnnUNet model to obtain segmented images of the apex myocardial region. Training 3DnnUNet models for nipple and basal muscle classes: Using a second 3DnnUNet architecture, network parameters are set; the model is trained using the cross-entropy loss function and the Adam optimizer, with data augmentation performed during training to increase the model's generalization ability; After training, the optimal model weights are saved, and the delayed-enhanced cardiac CT images of the papillary muscle category and the basal category are input into the optimal 3DnnUNet model to obtain segmented images of the papillary muscle myocardial region and the basal myocardial region. The segmentation results of the left ventricular myocardial region are obtained by stitching together the original DICOM images of the apex, papillary muscles, and base of the heart with the corresponding myocardial region segmentation labels of these three images.

2. The method for obtaining segmental myocardial CT values ​​of the left ventricle according to claim 1, characterized in that, The data used to train the 3DnnUNet model for the apex category is delayed-enhanced cardiac CT images of the apex category and their corresponding labels; The data used to train the 3DnnUNet models for the papillary muscle and basal muscle categories are delayed-enhanced cardiac CT images of the papillary muscle category, delayed-enhanced cardiac CT images of the basal muscle category, and their corresponding labels.

3. The method for obtaining segmental myocardial CT values ​​of the left ventricle according to claim 1, characterized in that, The specific steps for further subdividing the segmented left ventricular myocardial region are as follows: The pydicom.dcmread function is used to read three DICOM images; the nib.load function is used to read the segmentation mask image label corresponding to each image; the three DICOM images are the most representative images of the apex, papillary muscle, and basal categories; Adjust the labels of the read DICOM image and segmentation mask image so that their positions correspond to each other; Traverse each segmentation mask image label and record the positions where the pixel value is 0; also set the pixel value of the corresponding pixel in the DICOM image to 0, thereby determining the region on the DICOM image where the CT value needs to be calculated; Cropping out the excess black background from the image and reshaping the cropped image into a square; The function `get6.py` and `get4.py` retrieves two regions. In `get6.py`, the delayed-enhanced CT image of the basal and papillary muscles is divided into six parts according to a 16-segment model. It first defines the vertex coordinates of six fan-shaped regions, then segments the input image into six parts and stores them in the `segments` list. Similarly, in `get4.py`, the delayed-enhanced CT image of the apex is divided into four parts according to a 16-segment model. It first defines the vertex coordinates of four fan-shaped regions, then segments the input image into four parts and stores them in the `segments` list.

4. The method for obtaining segmental myocardial CT values ​​of the left ventricle according to claim 1, characterized in that, The method for converting pixel mean to the corresponding CT mean is as follows: in, For delayed enhancement images of the heart CT scan, For the slope of delayed-contrast cardiac CT images, This is the intercept.

5. The method for obtaining segmental myocardial CT values ​​of the left ventricle according to claim 4, characterized in that, The pixel mean of each region is obtained and then converted into the corresponding CT mean, specifically as follows: Obtain the rescale intercept and rescale slope for each DICOM image; Iterate through the 6 regions of the basal CT delayed-enhanced image and the 6 regions of the papillary muscle CT delayed-enhanced image output by the get6.py function, read the pixel value of each pixel in each region, and then obtain the pixel mean. Use a conversion formula to convert the pixel mean of each region to the CT mean. Iterate through the 4 regions of the apex CT delayed-enhanced image output by the get4.py function, read the pixel value of each pixel in each region, and then obtain the pixel mean. Use a conversion formula to convert the pixel mean of each region to the CT mean. The average CT values ​​of 6 regions from the delayed enhancement images of the basal CT, 6 regions from the delayed enhancement images of the papillary muscle CT, and 4 regions from the delayed enhancement images of the apex CT are stored in the list averagesCT. The mean CT values ​​were mapped to 16 regions, and a 16-segment model was drawn. The segmented mask images of each myocardial region are stacked together.

6. A system for acquiring segmental myocardial CT values ​​of the left ventricle, characterized in that, include: A dataset module was constructed to perform delayed-enhanced CT scans of the left ventricle of multiple patients to obtain complete left ventricular image data; different algorithms were used to slice and reconstruct the left ventricular image data to obtain delayed-enhanced cardiac CT images for each patient; The image preprocessing module preprocesses the delayed-enhancement cardiac CT image, including image normalization and noise removal, to improve image quality. The myocardial region segmentation module uses the 3DnnUNet model to segment the preprocessed cardiac CT delayed enhancement image to obtain the segmentation result of the left ventricular myocardial region. The specific implementation process is as follows: The preprocessed cardiac CT delayed enhancement images were divided into apex, papillary muscle, and basal categories; Training the 3DnnUNet model for the apex category: Using the first 3DnnUNet architecture, network parameters are set; the model is trained using the cross-entropy loss function and the Adam optimizer, and data augmentation is performed during training to increase the model's generalization ability; After training, the optimal model weights are saved, and the delayed-enhanced cardiac CT images of the apex category are input into the optimal 3DnnUNet model to obtain segmented images of the apex myocardial region. Training 3DnnUNet models for nipple and basal muscle classes: Using a second 3DnnUNet architecture, network parameters are set; the model is trained using the cross-entropy loss function and the Adam optimizer, with data augmentation performed during training to increase the model's generalization ability; After training, the optimal model weights are saved, and the delayed-enhanced cardiac CT images of the papillary muscle category and the basal category are input into the optimal 3DnnUNet model to obtain segmented images of the papillary muscle myocardial region and the basal myocardial region. The segmentation result of the left ventricular myocardial region is obtained by stitching together the original DICOM images of the apex, papillary muscles, and base of the heart with the corresponding myocardial region segmentation labels of these three images. The region segmentation module further subdivides the segmented left ventricular myocardial region: the apical CT delayed enhancement image is divided into 4 regions: the anterior wall of the apex, the apical septum, the inferior wall of the apex, and the lateral wall of the apex; the papillary muscle CT delayed enhancement image is divided into 6 regions: the anterior wall of the middle part, the anterior septum of the middle part, the inferior septum of the middle part, the inferior wall of the middle part, the inferior lateral wall of the middle part, and the anterior lateral wall of the middle part; the basal CT delayed enhancement image is divided into 6 regions: the anterior wall of the basement, the anterior septum of the basement, the inferior septum of the basement, the inferior wall of the basement, the inferior lateral wall of the basement, and the anterior lateral wall of the basal segment. The CT value acquisition module calculates the CT value for each region: it obtains the pixel mean for each region and then converts it into the corresponding CT mean.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running thereon, characterized in that, When the processor executes the program, it implements the method for obtaining CT values ​​of left ventricular segmental myocardium as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements a method for obtaining CT values ​​of left ventricular segmental myocardium as described in any one of claims 1-5.

Images