Image automatic measurement method and device, equipment and storage medium

Abstract

The invention relates to the technical field of image processing, and discloses an automatic image measurement method and device, equipment and a storage medium. The method comprises the following steps: firstly, segmenting a to-be-detected three-dimensional medical image formed by stacking a plurality of two-dimensional slice images, and determining a target area comprising a left ventricular area and a right ventricular area in each two-dimensional slice image; then, based on the target area in each two-dimensional slice image, determining left and right ventricle segmentation lines in each two-dimensional slice image; and finally, measuring the left ventricular area and the right ventricular area based on the left and right ventricular segmentation lines in each two-dimensional slice image to obtain a measurement result comprising at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricular diameter to the left ventricular diameter, so that an automatic process can be realized; and a rapid and accurate measurement result is provided, errors occurring during manual measurement are greatly reduced, and the influence of human intervention and individual deviation on the measurement result is avoided, so that the measurement efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an automatic image measurement method, device, equipment and storage medium. Background Art

[0002] Background: Right ventricular (RV) dysfunction caused by acute pulmonary embolism (PE) is associated with poor prognosis, and RV dilatation can be used as a surrogate marker of dysfunction, which is assessed by calculating the right ventricular to left ventricular diameter ratio (RV / LV) on standard CTPA images.

[0003] In related technologies, doctors need to identify the short-axis diameters of the left and right ventricles in cardiac images layer by layer, find the maximum diameter and calculate the RV / LV ratio. However, due to individual differences and the workload of multi-layer measurement, the accuracy of the results needs to be improved. Summary of the invention

[0004] The embodiments of this specification are intended to solve at least one of the technical problems in the related art to a certain extent. To this end, the embodiments of this specification propose an automatic image measurement method, device, equipment and storage medium.

[0005] The present specification provides an automatic image measurement method, the method comprising:

[0006] Segmenting the three-dimensional medical image to be detected, and determining a target area in each two-dimensional slice image, wherein the three-dimensional medical image to be detected is formed by stacking multiple two-dimensional slice images, and the target area includes a left ventricle area and a right ventricle area;

[0007] Based on the target area in each of the two-dimensional slice images, determining a left and right ventricle segmentation line in each of the two-dimensional slice images;

[0008] The left ventricle region and the right ventricle region are measured based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain measurement results, wherein the measurement results include at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

[0009] In one embodiment, the target region further includes a left ventricular myocardial region; and determining the left and right ventricle segmentation lines in each two-dimensional slice image based on the target region in each two-dimensional slice image includes:

[0010] Based on the target area in each two-dimensional slice image, determining the angle of the long axis direction of the heart in each two-dimensional slice image;

[0011] A straight line along the angle of the long axis of the heart and passing through the center of the interventricular septum region is used as the left and right ventricle dividing line, wherein the interventricular septum region is a partial region of the left ventricular myocardial region.

[0012] In one embodiment, the ventricular septum region is determined by:

[0013] Based on the left ventricular region and the right ventricular region in each two-dimensional slice image, determining the centroid of the left ventricular region and the centroid of the right ventricular region;

[0014] Based on a straight line parallel to a line connecting the center of mass of the left ventricular region and the center of mass of the right ventricular region, the left ventricular myocardial region satisfying a preset ratio is determined as the interventricular septum region.

[0015] In one embodiment, determining the angle of the long axis direction of the heart in each two-dimensional slice image based on the target area in each two-dimensional slice image includes:

[0016] Based on the left ventricular region in each two-dimensional slice image, determining the centroid of the left ventricular region;

[0017] Based on the centroid of the left ventricular region, the second-order image moment of the left ventricular region is calculated, so as to determine the angle of the long axis direction of the heart in each two-dimensional slice image.

[0018] In one embodiment, determining the angle of the long axis direction of the heart in each two-dimensional slice image based on the target area in each two-dimensional slice image includes:

[0019] Performing image processing on each two-dimensional slice image to determine the ventricular septum region in each two-dimensional slice image;

[0020] Based on the ventricular septum region in each two-dimensional slice image, determining the centroid of the ventricular septum region;

[0021] Based on the centroid of the interventricular septum region, the second-order image moment of the interventricular septum region is calculated, so as to determine the angle of the long axis direction of the heart in each two-dimensional slice image.

[0022] In one embodiment, determining the maximum left ventricular internal diameter based on the left and right ventricle dividing lines and the left ventricular region in each two-dimensional slice image includes:

[0023] On each of the two-dimensional slice images, the straight line perpendicular to the left and right ventricle dividing line is moved from one side of the left ventricle region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the left ventricle region, and the longest line segment is used as the maximum left ventricular inner diameter.

[0024] In one embodiment, determining the maximum right ventricular internal diameter based on the left and right ventricle dividing lines and the right ventricular region in each two-dimensional slice image includes:

[0025] On each of the two-dimensional slice images, the straight line perpendicular to the left and right ventricle dividing line is moved from one side of the right ventricle region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the right ventricle region, and the longest line segment is used as the maximum right ventricular inner diameter.

[0026] In one embodiment, the method further comprises:

[0027] Displaying the corresponding right ventricular internal diameter and left ventricular internal diameter in each two-dimensional slice image, wherein the multiple right ventricular internal diameters include the maximum right ventricular internal diameter and other right ventricular internal diameters, the multiple left ventricular internal diameters include the maximum left ventricular internal diameter and other left ventricular internal diameters, the display mode of the maximum right ventricular internal diameter is different from the display mode of the other right ventricular internal diameters, and the display mode of the maximum left ventricular internal diameter is different from the display mode of the other left ventricular internal diameters;

[0028] The measurement result is re-determined when at least one of the plurality of right ventricular internal diameters and / or at least one of the plurality of left ventricular internal diameters is adjusted.

[0029] The present specification provides an automatic image measurement device, the device comprising:

[0030] A target region determination module, used for segmenting the three-dimensional medical image to be detected and determining a target region in each two-dimensional slice image, wherein the three-dimensional medical image to be detected is formed by stacking a plurality of two-dimensional slice images, and the target region includes a left ventricle region and a right ventricle region;

[0031] A left-right ventricle segmentation line determination module, configured to determine a left-right ventricle segmentation line in each two-dimensional slice image based on the target area in each two-dimensional slice image;

[0032] A measurement result generating module is used to measure the left ventricle region and the right ventricle region based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain measurement results, wherein the measurement results include at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

[0033] An embodiment of the present specification provides a medical imaging device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method described in any of the above embodiments.

[0034] An embodiment of the present specification provides a computer device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method described in any of the above embodiments.

[0035] The embodiments of this specification provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0036] An embodiment of the present specification provides a computer program product, wherein the computer program product includes instructions, and when the instructions are executed by a processor of a computer device, the computer device is enabled to perform the steps of the method described in any one of the above embodiments.

[0037] In the above-mentioned implementation of the specification, first, the three-dimensional medical image to be detected, which is formed by stacking multiple two-dimensional slice images, is segmented, and the target area including the left ventricle area and the right ventricle area is determined in each two-dimensional slice image. Then, based on the target area in each two-dimensional slice image, the left and right ventricle dividing lines in each two-dimensional slice image are determined. Finally, the left ventricle area and the right ventricle area are measured based on the left and right ventricle dividing lines in each two-dimensional slice image, and at least one of the measurement results including the maximum left ventricular internal diameter, the maximum right ventricular internal diameter and the ratio of the right ventricle to the left ventricle diameter is obtained. Through the above-mentioned implementation, an automated process can be realized, and fast and accurate measurement results can be provided, which greatly reduces the errors occurring during manual measurement, avoids the influence of human intervention and individual deviation on the measurement results, and thus improves the measurement efficiency. In addition, due to the complex structure of the heart ventricle, the above-mentioned implementation can provide the measurement results of layer-by-layer images, ensuring the continuity and reliability of the layer-by-layer measurement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram of the process flow of the automatic image measurement method provided in the embodiment of this specification;

[0039] Figure 2 A schematic diagram of a flow chart for determining the left and right ventricle dividing line provided in an embodiment of this specification;

[0040] Figure 3a A schematic diagram of a flow chart for determining a ventricular septum region provided for an embodiment of this specification;

[0041] Figure 3b A schematic diagram of the interventricular septum region provided for embodiments of the present specification;

[0042] Figure 4 A schematic diagram of a process for determining the angle of the long axis direction of the heart in each two-dimensional slice image provided in an embodiment of this specification;

[0043] Figure 5 A schematic diagram of a process for determining the angle of the long axis direction of the heart in each two-dimensional slice image provided in an embodiment of this specification;

[0044] Figure 6 A schematic diagram of determining the left ventricular internal diameter provided for an embodiment of this specification;

[0045] Figure 7 A schematic diagram of determining the right ventricular internal diameter provided for an embodiment of this specification;

[0046] Figure 8a A schematic diagram of a process for re-determining a measurement result provided in an embodiment of this specification;

[0047] Figure 8b A schematic diagram of the maximum left ventricular internal diameter and the maximum right ventricular internal diameter provided for an embodiment of this specification;

[0048] Fig. 9 A schematic diagram of an automatic image measurement device provided in an embodiment of this specification;

[0049] Fig.10 An internal structural diagram of a computer device provided for an embodiment of this specification. DETAILED DESCRIPTION

[0050] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0051] Right ventricular (RV) dysfunction caused by acute pulmonary embolism (PE) is closely associated with poor short-term and long-term prognosis. RV dilatation is an important indicator of RV dysfunction and can be assessed by calculating the right ventricular to left ventricular diameter ratio (RV / LV) on standard computed tomography pulmonary angiography (CTPA) images. As CTPA is commonly used to diagnose acute PE, it is a simple and economical option to evaluate cardiac function at the time of pulmonary embolism diagnosis and to assess long-term prognosis in outpatients.

[0052] Specifically, first, the diameter of the ventricle is measured in a standard axial view. Next, the maximum distance between the ventricular wall endocardium and the interventricular septum is measured perpendicular to the long axis of the heart. Then, the maximum diameters of the right and left ventricles are measured at different levels. Finally, if the RV / LV diameter ratio is ≥1.0, right ventricular dilatation is defined.

[0053] When measuring the diameter of the left and right ventricles, doctors need to identify the short-axis diameters of the left and right ventricles layer by layer on multi-layer cardiac images and ensure that the maximum diameter of the two ventricles in each layer is measured. Since there may be individual differences in measurements between different doctors, and layer-by-layer measurement of multi-layer images will increase the workload, the accuracy of the results needs to be improved.

[0054] Based on this, the embodiment of this specification provides an automatic image measurement method. First, the three-dimensional medical image to be detected, which is formed by stacking multiple two-dimensional slice images, is segmented, and the target area including the left ventricle area and the right ventricle area is determined in each two-dimensional slice image. Then, based on the target area in each two-dimensional slice image, the left and right ventricle dividing lines in each two-dimensional slice image are determined. Finally, the left ventricle area and the right ventricle area are measured based on the left and right ventricle dividing lines in each two-dimensional slice image, and the measurement results including at least one of the maximum left ventricular internal diameter, the maximum right ventricular internal diameter and the ratio of the right ventricle to the left ventricle diameter are obtained. Through the above embodiment, an automated process can be realized, and fast and accurate measurement results can be provided, which greatly reduces the errors occurring during manual measurement, avoids the influence of human intervention and individual deviation on the measurement results, and thus improves the measurement efficiency. In addition, due to the complex structure of the heart ventricle, the above embodiment can provide the measurement results of layer-by-layer images, ensuring the continuity and reliability of the layer-by-layer measurement results.

[0055] The embodiment of this specification provides an application scenario of an automatic image measurement method. Specifically, the CTPA image to be detected is input into the segmentation model, and the left ventricle, right ventricle and left ventricular myocardium are segmented. The left ventricle region, right ventricle region and left ventricular myocardium region are determined in each two-dimensional slice image, wherein the CTPA image to be detected is formed by stacking multiple two-dimensional slice images. Then, a two-dimensional slice image is selected along the axial direction of the CTPA image to be detected, the angle of the long axis direction of the heart in the two-dimensional slice image is determined, and the straight line along the angle of the long axis direction of the heart and passing through the center of the ventricular septum region is used as the left and right ventricle segmentation line. Then, on the two-dimensional slice image, based on the left and right ventricle segmentation lines, the left ventricle region and the right ventricle region in the two-dimensional slice image, the left ventricle inner diameter and the right ventricle inner diameter of the two-dimensional slice image are determined and the inner diameter line segment is recorded. Then, it is determined whether the last two-dimensional slice image has been traversed. If the last two-dimensional slice image has not been traversed, another two-dimensional slice image is selected to repeat the above operation. If the last two-dimensional slice image is traversed, the maximum left ventricular internal diameter and the maximum right ventricular internal diameter are determined, and then the ratio of the right ventricle to the left ventricle diameter is calculated based on the maximum left ventricular internal diameter and the maximum right ventricular internal diameter. Finally, the maximum left ventricular internal diameter, the maximum right ventricular internal diameter and the ratio of the right ventricle to the left ventricle diameter are output, and the left ventricular internal diameter, the right ventricular internal diameter and the left and right ventricle dividing line of the current two-dimensional slice image are marked on each two-dimensional slice image of the CTPA image to be detected.

[0056] This specification provides an automatic image measurement method. Figure 1 , the image automatic measurement method may include the following steps:

[0057] S110 , segmenting the three-dimensional medical image to be detected, and determining a target area in each two-dimensional slice image.

[0058] The three-dimensional medical image to be detected is formed by stacking multiple two-dimensional slice images, and the target area includes the left ventricle area and the right ventricle area.

[0059] Specifically, a medical imaging device is used to collect data and reconstruct images of a scanning area including a heart part to obtain a three-dimensional medical image to be detected, wherein the three-dimensional medical image to be detected is formed by stacking multiple two-dimensional slice images. Then, the three-dimensional medical image to be detected is segmented using computer vision technology to determine a target area including a left ventricle area and a right ventricle area in each two-dimensional slice image. Exemplarily, the three-dimensional medical image to be detected may be a CTPA image.

[0060] In some embodiments, the three-dimensional medical image to be detected may be subjected to image preprocessing (such as normalization) to obtain input data of the segmentation model. The segmentation model obtains the target region including the left ventricle region and the right ventricle region through forward propagation.

[0061] Exemplarily, segmenting the three-dimensional medical image to be detected can be implemented using a deep learning network, for example, a network based on a U-Net structure. The specific processing flow may include data preprocessing, model building, and model training.

[0062] Data preprocessing: First, collect a CTPA image dataset containing annotations of the right ventricle, left ventricle, and left ventricular myocardium. The dataset should cover diverse samples from different patients to ensure the generalization ability of the model. Annotate the left ventricle region, right ventricle region, and left ventricular myocardium region in each CTPA image in the dataset. For example, each region in the CTPA image can be labeled with a different numerical value, for example, the right ventricle region is marked as 1, the left ventricle region is marked as 2, the left ventricular myocardium region is marked as 3, and the rest is regarded as the background and marked as 0. Then, the pixel value range of each CTPA image is normalized to the same scale, for example, the pixel values ​​of -1000 to 2000 are scaled to between 0 and 1. Next, since the spatial resolution of the image may be different, all CTPA images need to be resampled to ensure that each CTPA image has the same spatial resolution to maintain the applicability of the model to images of different scales. In addition, in order to increase the diversity of the data and reduce overfitting, random rotation, translation, scaling, cropping, and elastic deformation techniques can also be used for data enhancement. Finally, the annotations are converted into mask images with the same resolution as the image, corresponding to the right ventricle, left ventricle, and left ventricular myocardium. The annotation data is converted into mask images with the same resolution as the original image, corresponding to the left ventricle area, right ventricle area, and left ventricular myocardium area.

[0063] Model structure: The segmentation network is constructed based on the Unet-like network model structure. The model consists of an encoder, a decoder, and an output layer. The encoder is used to extract high-level features of the image from the input image. The encoder contains multiple convolutional blocks, each of which consists of a convolutional layer, a batch normalization layer, an activation layer (such as ReLU), and a pooling layer. Downsampling is performed through pooling layers (such as maximum pooling), gradually reducing the spatial size of the feature map and increasing the number of feature channels, thereby capturing high-level semantic features in the image.

[0064] The decoder is used to restore the spatial dimensions and details of the image. The decoder contains multiple upsampling layers (such as transposed convolution or bilinear interpolation), each of which is followed by a convolution block. The decoder fuses the feature maps in the contraction path with the feature maps in the expansion path through skip connections to preserve more detail information.

[0065] The output layer uses a convolutional layer to map the last feature map of the network to a segmentation mask map with the same resolution as the input image. The output segmentation mask will have three channels, corresponding to the segmented regions of the right ventricle, left ventricle, and left ventricular myocardium.

[0066] Training process: The pixel-level classification accuracy and the overlap of the segmented regions can be comprehensively considered, and the segmentation model can be optimized by using a combination of the cross entropy loss function and the Dice loss function. The optimizer can select the Adam optimizer for parameter optimization. Specifically, the CTPA image that has undergone data preprocessing and the corresponding annotated mask data are input into the segmentation model to be trained. Next, the model calculates the prediction result through forward propagation, and calculates the loss value based on the prediction result and the actual label. Then, the model parameters are adjusted using the back propagation algorithm to reduce the loss function value. Repeat the above process until the performance of the model on the validation set converges, that is, the loss value on the validation set no longer decreases significantly, or the predetermined training rounds are reached, thereby obtaining a trained segmentation model.

[0067] S120 , determining a left and right ventricle segmentation line in each two-dimensional slice image based on the target region in each two-dimensional slice image.

[0068] S130 , measuring the left ventricle region and the right ventricle region based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain a measurement result.

[0069] The measurement results include at least one of the maximum left ventricular internal diameter, the maximum right ventricular internal diameter, and the ratio of the right ventricle diameter to the left ventricle diameter.

[0070] Specifically, for the left ventricular region and the right ventricular region on each two-dimensional slice image, the boundaries of the left ventricular region and the right ventricular region are clearly marked through an automatic segmentation algorithm or manual annotation, and the left and right ventricular dividing lines in each two-dimensional slice image are determined. For each two-dimensional slice image, the left and right ventricular dividing lines are used as the boundaries, and the left ventricular region and the right ventricular region are measured to obtain the left ventricular inner diameter and the right ventricular inner diameter corresponding to the two-dimensional slice image. The left ventricular inner diameter and the right ventricular inner diameter corresponding to each two-dimensional slice image are compared, and the maximum value is taken and used as the maximum left ventricular inner diameter and the maximum right ventricular inner diameter. The maximum right ventricular inner diameter is divided by the maximum left ventricular inner diameter to obtain the ratio of the right ventricular diameter to the left ventricular diameter. At least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricular diameter to the left ventricular diameter constitutes the measurement result.

[0071] In the above implementation, first, the three-dimensional medical image to be detected, which is formed by stacking multiple two-dimensional slice images, is segmented, and the target area including the left ventricle area and the right ventricle area is determined in each two-dimensional slice image. Then, based on the target area in each two-dimensional slice image, the left and right ventricle dividing lines in each two-dimensional slice image are determined. Finally, the left ventricle area and the right ventricle area are measured based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain at least one of the measurement results including the maximum left ventricular internal diameter, the maximum right ventricular internal diameter and the ratio of the right ventricle to the left ventricle diameter. Through the above implementation, an automated process can be realized, and fast and accurate measurement results can be provided, which greatly reduces the errors occurring during manual measurement, avoids the influence of human intervention and individual deviation on the measurement results, and thus improves the measurement efficiency. In addition, due to the complex structure of the heart ventricle, the above implementation can provide the measurement results of layer-by-layer images, ensuring the continuity and reliability of the layer-by-layer measurement results.

[0072] In some embodiments, see Figure 2 , the target area also includes the left ventricular myocardial area; based on the target area in each two-dimensional slice image, determining the left and right ventricle segmentation line in each two-dimensional slice image may include the following steps:

[0073] S210 : Determine the angle of the long axis direction of the heart in each two-dimensional slice image based on the target area in each two-dimensional slice image.

[0074] S220, taking a straight line along the long axis of the heart and passing through the center of the interventricular septum region as a dividing line between the left and right ventricles.

[0075] The interventricular septum region is a part of the left ventricular myocardial region.

[0076] Specifically, based on the target area in any two-dimensional slice image, based on the geometric shape and anatomical structure of the heart, the image analysis technology is used to determine the angle of the long axis direction of the heart in each two-dimensional slice image. After determining the angle of the long axis direction of the heart, a segmentation line along the long axis direction of the heart is drawn, and passes through the center point of the ventricular septum area, because the ventricular septum area is usually located in the center of the heart and is an important mark for distinguishing the left and right ventricles, so that the dividing line is used as the left and right ventricle segmentation line. Repeat the above operation for each two-dimensional slice image to determine the left and right ventricle segmentation line in each two-dimensional slice image.

[0077] In some implementations, the angle of the long axis direction of the heart in the two-dimensional slice image may be determined by fitting the boundary between the left ventricle region and the right ventricle region in the two-dimensional slice image.

[0078] In the above implementation, based on the target area in each two-dimensional slice image, the angle of the long axis direction of the heart in each two-dimensional slice image is determined, and a straight line along the angle of the long axis direction of the heart and passing through the center of the interventricular septum area is used as the left and right ventricle dividing line, thereby providing a data basis for the subsequent determination of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

[0079] In some embodiments, see Figure 3a , the ventricular septal area is determined by:

[0080] S310 , based on the left ventricular region and the right ventricular region in each two-dimensional slice image, determine the center of mass of the left ventricular region and the center of mass of the right ventricular region.

[0081] S320: Based on a straight line parallel to a line connecting the center of mass of the left ventricular region and the center of mass of the right ventricular region, determine a left ventricular myocardial region that meets a preset ratio as a ventricular septum region.

[0082] Specifically, the coordinates of all pixels in the left ventricular region in any two-dimensional slice image are found, and then the average value of the coordinates of all these pixels is calculated to obtain the center of mass of the left ventricular region. The coordinates of all pixels in the right ventricular region in the two-dimensional slice image are found, and then the average value of the coordinates of all these pixels is calculated to obtain the center of mass of the right ventricular region. A straight line parallel to the line connecting the center of mass of the left ventricular region and the center of mass of the right ventricular region is moved from one side of the left and right ventricles to the other side. During this process, when the straight line passes through the left ventricular myocardial region, it is determined which parts of the left ventricular myocardial region occupy a proportion of the straight line between the left and right ventricles that meets a preset proportion (for example, 80%). When the left ventricular myocardial region on the straight line meets the preset proportion, the left ventricular myocardial region at that location is determined as the interventricular septum region. Finally, all left ventricular myocardial regions that meet the preset proportion constitute the interventricular septum region. Repeat the above operation for each two-dimensional slice image to determine the interventricular septum region in each two-dimensional slice image. For example, please refer to Figure 3b , Figure 3b The rectangular area 304 in the left ventricular myocardium area 302 is the interventricular septum area.

[0083] In the above implementation, based on the left ventricular region and the right ventricular region in each two-dimensional slice image, the center of mass of the left ventricular region and the center of mass of the right ventricular region are determined, and based on a straight line parallel to the line connecting the center of mass of the left ventricular region and the center of mass of the right ventricular region, the left ventricular myocardial region that meets a preset ratio is determined as the ventricular septum region, providing a data basis for the subsequent determination of the left and right ventricular dividing line.

[0084] In some embodiments, see Figure 4, based on the target area in each two-dimensional slice image, determining the angle of the long axis direction of the heart in each two-dimensional slice image may include the following steps:

[0085] S410 , determining the centroid of the left ventricular region based on the left ventricular region in each two-dimensional slice image.

[0086] S420. Calculate the second-order image moment of the left ventricular region based on the centroid of the left ventricular region, thereby determining the angle of the long axis direction of the heart in each two-dimensional slice image.

[0087] Specifically, the coordinates of all pixels in the left ventricular region in any two-dimensional slice image are found, and then the average value of the coordinates of all these pixels is calculated to obtain the centroid of the left ventricular region. Based on the centroid of the left ventricular region, the maximum second-order image central moment corresponding to the left ventricular region is calculated to determine the angle of the long axis direction of the heart in the two-dimensional slice image. The above operation is repeated for each two-dimensional slice image to determine the angle of the long axis direction of the heart in each two-dimensional slice image.

[0088] In the above implementation, the center of mass of the left ventricular region is determined based on the left ventricular region in each two-dimensional slice image, and the second-order image moment of the left ventricular region is calculated based on the center of mass of the left ventricular region, thereby determining the angle of the long axis direction of the heart in each two-dimensional slice image, providing a data basis for the subsequent determination of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricular to left ventricular diameters.

[0089] In some embodiments, see Figure 5 , based on the target area in each two-dimensional slice image, determining the angle of the long axis direction of the heart in each two-dimensional slice image may include the following steps:

[0090] S510 , performing image processing on each two-dimensional slice image to determine the ventricular septum region in each two-dimensional slice image.

[0091] S520 , determining the centroid of the ventricular septum region based on the ventricular septum region in each two-dimensional slice image.

[0092] S530 , calculating the second-order image moment of the ventricular septum region based on the centroid of the ventricular septum region, thereby determining the angle of the long axis direction of the heart in each two-dimensional slice image.

[0093] Specifically, considering the irregularity of the shape of the ventricular region, the calculation of the long axis direction based on the left ventricle will be affected by the shape of the left ventricle, resulting in instability of the long axis direction. First, each two-dimensional slice image is processed to determine the center of mass of the left ventricular region and the center of mass of the right ventricular region. Then, based on the straight line parallel to the line connecting the center of mass of the left ventricular region and the center of mass of the right ventricular region, the left ventricular myocardial region that meets the preset ratio is determined as the ventricular septum region. By finding the coordinates of all pixel points in the ventricular septum region in any two-dimensional slice image, and then calculating the average value of the coordinates of all these pixel points, the center of mass of the ventricular septum region is obtained. Based on the center of mass of the ventricular septum region, the maximum second-order image center moment corresponding to the ventricular septum region is calculated to determine the angle of the long axis direction of the heart in the two-dimensional slice image. Repeat the above operation for each two-dimensional slice image to determine the angle of the long axis direction of the heart in each two-dimensional slice image to achieve a more stable long axis direction estimation.

[0094] In the above implementation, image processing is performed on each two-dimensional slice image to determine the ventricular septum region in each two-dimensional slice image. Based on the ventricular septum region in each two-dimensional slice image, the center of mass of the ventricular septum region is determined. Based on the center of mass of the ventricular septum region, the second-order image moment of the ventricular septum region is calculated, thereby determining the angle of the long axis direction of the heart in each two-dimensional slice image, providing a data basis for the subsequent determination of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

[0095] In some embodiments, determining the maximum left ventricular internal diameter based on the left and right ventricle dividing line and the left ventricular region in each two-dimensional slice image may include: on each two-dimensional slice image, moving a straight line perpendicular to the left and right ventricle dividing line from one side of the left ventricular region to the other side based on a received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the left ventricular region, and using the longest line segment as the maximum left ventricular internal diameter.

[0096] Specifically, on each two-dimensional slice image, based on the received movement instruction, a straight line perpendicular to the left and right ventricle dividing line is moved from one side boundary of the left ventricle region as a starting point along the long axis of the heart at a predetermined sampling interval and sampled. At each sampling position, the length and position of the longest intersection line segment of the straight line perpendicular to the left and right ventricle dividing line and the left ventricle region are calculated and recorded as the left ventricular internal diameter of the two-dimensional slice image. The left ventricular internal diameters in each two-dimensional slice image are compared, and the maximum value, i.e., the longest line segment, is taken as the maximum left ventricular internal diameter.

[0097] For example, see Figure 6For any two-dimensional slice image, based on the received movement instruction, the straight line B perpendicular to the left and right ventricle dividing line A is moved from one side of the left ventricle region to the other side at n (for example, n=1) pixel intervals, and the longest intersection area line segment D between the straight line B and the left ventricle region is recorded. The longest intersection area line segment D is the left ventricle inner diameter of the two-dimensional slice image.

[0098] In the above embodiment, on each two-dimensional slice image, the straight line perpendicular to the left and right ventricle dividing line is moved from one side of the left ventricle region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the left ventricle region, and the longest line segment is used as the maximum left ventricular inner diameter, providing a data basis for the subsequent determination of the ratio of the right ventricle to the left ventricle diameter.

[0099] In some embodiments, determining the maximum right ventricular inner diameter based on the left and right ventricle dividing line and the right ventricular region in each two-dimensional slice image may include: on each two-dimensional slice image, moving the straight line perpendicular to the left and right ventricle dividing line from one side of the right ventricular region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the right ventricular region, and using the longest line segment as the maximum right ventricular inner diameter.

[0100] Specifically, on each two-dimensional slice image, based on the received movement instruction, a straight line perpendicular to the dividing line between the left and right ventricles is moved from one side boundary of the right ventricle region as a starting point along the long axis of the heart at a predetermined sampling interval and sampled. At each sampling position, the length and position of the longest intersection line segment of the straight line perpendicular to the dividing line between the left and right ventricles and the right ventricle region are calculated and recorded as the right ventricular internal diameter of the two-dimensional slice image. The right ventricular internal diameters in each two-dimensional slice image are compared, and the maximum value, i.e., the longest line segment, is taken as the maximum right ventricular internal diameter.

[0101] For example, see Figure 7 For any two-dimensional slice image, based on the received movement instruction, the straight line B perpendicular to the left and right ventricle dividing line A is moved from one side of the right ventricle region to the other side at n (for example, n=1) pixel intervals, and the longest intersection area line segment C between the straight line B and the right ventricle region is recorded. The longest intersection area line segment C is the right ventricle inner diameter of the two-dimensional slice image.

[0102] In the above embodiment, on each two-dimensional slice image, the straight line perpendicular to the left and right ventricle dividing line is moved from one side of the right ventricle region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the right ventricle region, and the longest line segment is used as the maximum right ventricular inner diameter, providing a data basis for the subsequent determination of the ratio of the right ventricle to the left ventricle diameter.

[0103] In some embodiments, see Figure 8a , the method may further include the following steps:

[0104] S810. Display the corresponding right ventricular inner diameter and left ventricular inner diameter in each two-dimensional slice image.

[0105] Among them, multiple right ventricular internal diameters include the maximum right ventricular internal diameter and other right ventricular internal diameters, and multiple left ventricular internal diameters include the maximum left ventricular internal diameter and other left ventricular internal diameters. The display method of the maximum right ventricular internal diameter is different from the display method of other right ventricular internal diameters, and the display method of the maximum left ventricular internal diameter is different from the display method of other left ventricular internal diameters.

[0106] Specifically, the corresponding right ventricular internal diameter and left ventricular internal diameter are displayed in each two-dimensional slice image. For the maximum right ventricular internal diameter, its display mode in the corresponding two-dimensional slice image is different from the display mode of other right ventricular internal diameters in the corresponding two-dimensional slice image. For the maximum left ventricular internal diameter, its display mode in the corresponding two-dimensional slice image is different from the display mode of other left ventricular internal diameters in the corresponding two-dimensional slice image.

[0107] For example, see Figure 8b , Figure 8b Shown in the figure are the maximum left ventricular internal diameter 802 and the maximum right ventricular internal diameter 804.

[0108] S820: When at least one of the plurality of right ventricular internal diameters and / or at least one of the plurality of left ventricular internal diameters is adjusted, re-determine the measurement result.

[0109] Specifically, after comprehensive observation and analysis of multiple right ventricular inner diameters and multiple left ventricular inner diameters, it is determined that the measurement of some inner diameters is inaccurate. Therefore, corresponding modifications and adjustments are required. First, at least one of the multiple right ventricular inner diameters and / or at least one of the multiple left ventricular inner diameters is identified as an adjustment target. Then, the inner diameter line segment is adjusted for the identified adjustment target, and the measurement result is re-determined. The measurement result includes at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter, and the ratio of the right ventricle to the left ventricle diameter.

[0110] In the above embodiment, corresponding multiple right ventricular internal diameters and multiple left ventricular internal diameters are displayed in each two-dimensional slice image, which is convenient for intuitive observation and confirmation. When at least one of the multiple right ventricular internal diameters and / or at least one of the multiple left ventricular internal diameters is adjusted, the measurement result is re-determined to determine the accuracy and reliability of the measurement result.

[0111] This specification provides an automatic image measurement device 900. Fig. 9The automatic image measurement device 900 includes: a target area determination module 910, a left and right ventricle dividing line determination module 920, and a measurement result generation module 930.

[0112] A target region determination module 910 is used to segment the three-dimensional medical image to be detected and determine a target region in each two-dimensional slice image, wherein the three-dimensional medical image to be detected is formed by stacking multiple two-dimensional slice images, and the target region includes a left ventricle region and a right ventricle region;

[0113] A left-right ventricle segmentation line determination module 920, configured to determine a left-right ventricle segmentation line in each two-dimensional slice image based on the target region in each two-dimensional slice image;

[0114] The measurement result generating module 930 is used to measure the left ventricle region and the right ventricle region based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain measurement results, wherein the measurement results include at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

[0115] For a detailed description of the automatic image measurement device, reference may be made to the above description of the automatic image measurement method, which will not be repeated here.

[0116] An embodiment of the present specification provides a medical imaging device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the method steps in the above embodiment when executing the computer program.

[0117] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.10 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an automatic image measurement method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0118] Those skilled in the art will understand that Fig.10 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution disclosed in this specification, and does not constitute a limitation on the computer device to which the solution disclosed in this specification is applied. Specifically, the computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] In some embodiments, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the method steps in the above embodiments when executing the computer program.

[0120] An embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method in any of the above embodiments are implemented.

[0121] One embodiment of the present specification provides a computer program product, which includes instructions. When the instructions are executed by a processor of a computer device, the computer device can perform the steps of the method of any of the above embodiments.

[0122] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

Claims

1. An automatic image measurement method, characterized in that: The method comprises: Segmenting the three-dimensional medical image to be detected, and determining a target area in each two-dimensional slice image, wherein the three-dimensional medical image to be detected is formed by stacking multiple two-dimensional slice images, and the target area includes a left ventricle area and a right ventricle area; Based on the target area in each of the two-dimensional slice images, determining a left and right ventricle segmentation line in each of the two-dimensional slice images; The left ventricle region and the right ventricle region are measured based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain measurement results, wherein the measurement results include at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

2. The method according to claim 1, characterized in that The target region also includes a left ventricular myocardial region; and determining the left and right ventricular segmentation lines in each two-dimensional slice image based on the target region in each two-dimensional slice image includes: Based on the target area in each two-dimensional slice image, determining the angle of the long axis direction of the heart in each two-dimensional slice image; A straight line along the angle of the long axis of the heart and passing through the center of the interventricular septum region is used as the left and right ventricle dividing line, wherein the interventricular septum region is a partial region of the left ventricular myocardial region.

3. The method according to claim 2, characterized in that The septal area is determined by: Based on the left ventricular region and the right ventricular region in each two-dimensional slice image, determining the centroid of the left ventricular region and the centroid of the right ventricular region; Based on a straight line parallel to a line connecting the center of mass of the left ventricular region and the center of mass of the right ventricular region, the left ventricular myocardial region satisfying a preset ratio is determined as the interventricular septum region.

4. The method according to claim 2 or 3, characterized in that: The step of determining the angle of the long axis direction of the heart in each two-dimensional slice image based on the target area in each two-dimensional slice image comprises: Based on the left ventricular region in each two-dimensional slice image, determining the centroid of the left ventricular region; Based on the centroid of the left ventricular region, the second-order image moment of the left ventricular region is calculated, so as to determine the angle of the long axis direction of the heart in each two-dimensional slice image.

5. The method according to claim 2 or 3, characterized in that: The step of determining the angle of the long axis direction of the heart in each two-dimensional slice image based on the target area in each two-dimensional slice image comprises: Performing image processing on each two-dimensional slice image to determine the ventricular septum region in each two-dimensional slice image; Based on the ventricular septum region in each two-dimensional slice image, determining the centroid of the ventricular septum region; Based on the centroid of the interventricular septum region, the second-order image moment of the interventricular septum region is calculated, so as to determine the angle of the long axis direction of the heart in each two-dimensional slice image.

6. The method according to any one of claims 1 to 3, characterized in that: Determining the maximum left ventricular internal diameter based on the left and right ventricle segmentation lines and the left ventricular region in each two-dimensional slice image includes: On each of the two-dimensional slice images, the straight line perpendicular to the left and right ventricle dividing line is moved from one side of the left ventricle region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the left ventricle region, and the longest line segment is used as the maximum left ventricular inner diameter.

7. The method according to any one of claims 1 to 3, characterized in that: Determining the maximum right ventricular internal diameter based on the left and right ventricle segmentation lines and the right ventricular region in each two-dimensional slice image includes: On each of the two-dimensional slice images, the straight line perpendicular to the left and right ventricle dividing line is moved from one side of the right ventricle region to the other side based on the received movement instruction to determine the longest line segment where the straight line perpendicular to the left and right ventricle dividing line intersects the right ventricle region, and the longest line segment is used as the maximum right ventricular inner diameter.

8. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: Displaying the corresponding right ventricular internal diameter and left ventricular internal diameter in each two-dimensional slice image, wherein the multiple right ventricular internal diameters include the maximum right ventricular internal diameter and other right ventricular internal diameters, the multiple left ventricular internal diameters include the maximum left ventricular internal diameter and other left ventricular internal diameters, the display mode of the maximum right ventricular internal diameter is different from the display mode of the other right ventricular internal diameters, and the display mode of the maximum left ventricular internal diameter is different from the display mode of the other left ventricular internal diameters; The measurement result is re-determined when at least one of the plurality of right ventricular internal diameters and / or at least one of the plurality of left ventricular internal diameters is adjusted.

9. An automatic image measuring device, characterized in that: The device comprises: A target region determination module, used for segmenting the three-dimensional medical image to be detected and determining a target region in each two-dimensional slice image, wherein the three-dimensional medical image to be detected is formed by stacking a plurality of two-dimensional slice images, and the target region includes a left ventricle region and a right ventricle region; A left-right ventricle segmentation line determination module, configured to determine a left-right ventricle segmentation line in each two-dimensional slice image based on the target area in each two-dimensional slice image; A measurement result generating module is used to measure the left ventricle region and the right ventricle region based on the left and right ventricle dividing lines in each two-dimensional slice image to obtain measurement results, wherein the measurement results include at least one of the maximum left ventricular inner diameter, the maximum right ventricular inner diameter and the ratio of the right ventricle to the left ventricle diameter.

10. A medical imaging device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

11. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images