Medical image detection method, device, storage medium and computer equipment

Abstract

The present application provides a medical image detection method, device, storage medium and computer equipment, the method comprising: acquiring a lung medical image; extracting arteries and bronchi in the lung medical image, wherein the bronchi include multiple bronchial segments of different levels; determining the target arterial segment corresponding to the bronchial segment in the lung medical image according to the distance between the centerline of the bronchial segment and the centerline of the arterial segment; detecting and processing the bronchial segment and the target arterial segment to determine the detection result of the bronchial segment. The method of the present application can automatically detect the columnar signs of the bronchi, which is conducive to improving the efficiency of doctors' film reading and reducing their workload.

Description

Technical Field

[0001] The present application relates to the technical field of medical image processing, and in particular to a medical image detection method, device, storage medium and computer equipment. Background Art

[0002] Bronchiectasis (abbreviated as bronchiectasis) is caused by chronic suppurative inflammation and fibrosis of the bronchi and surrounding lung tissue, which destroys the muscle and elastic tissue of the bronchial wall, leading to bronchial deformation and persistent dilatation. The imaging manifestations of bronchiectasis can be divided into columnar, cystic columnar, and cystic dilatation. For the same bronchial segment, columnar bronchiectasis is mostly a sign of the early development of the disease, while cystic bronchiectasis is a sign of the late stage of the disease, and cystic columnar bronchiectasis is somewhere in between. However, for different lobes of the lung, the degree of imaging involvement of bronchiectasis varies greatly from individual to individual.

[0003] The scoring system in the related technology can comprehensively evaluate bronchiectasis and its accompanying imaging signs. The system does not differentiate between the three signs in terms of process and method during the model training and use stages. However, since the cystic and cystic columnar signs are easy to distinguish from normal bronchi in terms of imaging characteristics, while the columnar signs are difficult to distinguish from normal bronchi, and the degree of imaging involvement of bronchiectasis varies greatly from individual to individual. If the same method is used to detect the three signs at the same time, the lack of consideration of a single individual factor will inevitably lead to low accuracy in detecting columnar signs, making it impossible to achieve a graded assessment of bronchiectasis. In this case, there will inevitably be large deviations and blank areas in the early diagnosis of bronchiectasis. Summary of the invention

[0004] In view of this, the present application provides a medical image detection method, device, storage medium and computer equipment, which can automatically detect the columnar signs of the bronchi, which is beneficial to improving the doctor's film reading efficiency and reducing the workload.

[0005] According to one aspect of the present application, a medical image detection method is provided, comprising:

[0006] Acquire a lung medical image; extract arteries and bronchi in the lung medical image, wherein the bronchi include multiple bronchial segments of different levels; determine a target arterial segment corresponding to the bronchial segment in the lung medical image based on a distance between a centerline of the bronchial segment and a centerline of the arterial vessel; perform detection and processing on the bronchial segment and the target arterial segment to determine a detection result of the bronchial segment.

[0007] Optionally, the bronchial segment and the target arterial segment are detected and processed to determine the detection result of the bronchial segment, specifically including:

[0008] According to a preset interval, the second bronchial segment in the lung medical image, at least one of the target arterial blood vessel segments corresponding to the first bronchial segment, and the first bronchial segment are intercepted and processed to obtain a first image, wherein the second bronchial segment is a bronchial segment of the previous level of the first bronchial segment; the first image and / or the mask of the first image are input into a columnar sign detection model to determine the detection result of the first bronchial segment; wherein the columnar sign detection model is trained based on a lung sample image and a columnar sign label corresponding to the lung sample image.

[0009] Optionally, the medical image detection method further includes:

[0010] A lung sample image is acquired, the lung sample image includes a plurality of bronchial segment samples of different levels and target arterial segment samples corresponding to the plurality of bronchial segment samples of different levels; when the diameter of the first bronchial segment sample is greater than the diameter of the corresponding target arterial segment sample, or the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is greater than a preset difference, a columnar sign label of the first bronchial segment sample is generated, the columnar sign label being used to mark the presence of a columnar sign in the bronchus; the second bronchial segment sample marked with the columnar sign label, at least one of the target arterial segment sample corresponding to the first bronchial segment sample, and the first bronchial segment sample are intercepted and processed to obtain a second image; a neural network model is trained according to the second image and / or the mask of the second image to obtain a columnar sign detection model.

[0011] Optionally, determining a target arterial blood vessel segment corresponding to the bronchial segment in the lung medical image according to a distance between a centerline of the bronchial segment and a centerline of the arterial blood vessel specifically includes:

[0012] Determine a first sampling point on the center line of the bronchial segment; calculate the distances between the first sampling point and multiple second sampling points on the center line of the arterial blood vessel respectively; determine the second sampling point corresponding to the smallest distance as the accompanying point corresponding to the first sampling point; determine at least one arterial blood vessel segment corresponding to the bronchial segment according to the starting point coordinates and the end point coordinates of the bronchial segment; select a target arterial blood vessel segment from at least one arterial blood vessel segment according to the position of the accompanying point.

[0013] Optionally, arteries and bronchi in the lung medical image are extracted, specifically including:

[0014] The lung medical image is input into the arterial segmentation model to obtain the arteries in the lung medical image; the lung medical image is input into the bronchial segmentation model to obtain the bronchi in the lung medical image; wherein the arterial segmentation model is trained based on the lung sample image and its corresponding arterial segmentation label, and the lung segmentation model is trained based on the lung sample image and its corresponding lung segmentation label.

[0015] Optionally, the medical image detection method further includes:

[0016] When the detection result shows that there is a columnar sign, the bronchial segment with the columnar sign in the lung medical image is differentially displayed.

[0017] Optionally, the medical image detection method further includes:

[0018] When the detection result shows that there is a columnar sign, the center point and the inner diameter of the bronchial segment where the columnar sign exists are associated and stored.

[0019] Optionally, obtain lung medical imaging, including:

[0020] Acquire a medical image to be detected; preprocess the medical image to be detected; perform lung segmentation processing on the preprocessed medical image to be detected to obtain a lung medical image; wherein the preprocessing includes at least one of the following: noise reduction processing, normalization processing, size correction processing, CT value truncation processing, and resampling processing.

[0021] According to another aspect of the present application, a medical image detection device is provided, the device comprising:

[0022] An acquisition module is used to acquire lung medical images; an extraction module is used to extract arteries and bronchi in lung medical images, wherein the bronchi include multiple bronchial segments of different levels; a determination module is used to determine the target arterial segment corresponding to the bronchial segment in the lung medical image based on the distance between the centerline of the bronchial segment and the centerline of the arterial segment; and a detection module is used to detect and process the bronchial segment and the target arterial segment to determine the detection result of the bronchial segment.

[0023] Optionally, the medical image detection device also includes: a capture module, which captures the second bronchial segment in the lung medical image, at least one of the target arterial blood vessel segments corresponding to the first bronchial segment, and the first bronchial segment according to a preset interval to obtain a first image, wherein the second bronchial segment is a bronchial segment of the previous level of the first bronchial segment; the detection module is specifically used to input the first image and / or the mask of the first image into a columnar sign detection model to determine the detection result of the first bronchial segment; wherein the columnar sign detection model is trained based on the lung sample image and the columnar sign label corresponding to the lung sample image.

[0024] Optionally, the acquisition module is also used to acquire a lung sample image, which includes multiple bronchial segment samples of different levels and target arterial segment samples corresponding to multiple bronchial segment samples of different levels; the medical image detection device also includes: a marking module, which is used to generate a columnar sign label for the first bronchial segment sample when the diameter of the first bronchial segment sample is larger than the diameter of the corresponding target arterial segment sample, or the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is larger than a preset difference, and the columnar sign label is used to mark the presence of a columnar sign in the bronchus; the interception module is also used to intercept the second bronchial segment sample marked with the columnar sign label, at least one of the target arterial segment sample corresponding to the first bronchial segment sample, and the first bronchial segment sample to obtain a second image; the medical image detection device also includes: a training module, which is used to train a neural network model based on the second image and / or the mask of the second image to obtain a columnar sign detection model.

[0025] Optionally, the determination module is specifically used to determine a first sampling point on the center line of the bronchial segment; respectively calculate the distances between the first sampling point and multiple second sampling points on the center line of the arterial blood vessel; the determination module is specifically used to determine the second sampling point corresponding to the smallest distance as the accompanying point corresponding to the first sampling point; determine at least one arterial blood vessel segment corresponding to the bronchial segment according to the starting point coordinates and the end point coordinates of the bronchial segment; and select a target arterial blood vessel segment from at least one arterial blood vessel segment according to the position of the accompanying point.

[0026] Optionally, the extraction module is specifically used to input the lung medical image into the arterial segmentation model to obtain the arteries in the lung medical image; input the lung medical image into the bronchial segmentation model to obtain the bronchi in the lung medical image; wherein the arterial segmentation model is trained based on the lung sample image and its corresponding arterial segmentation label, and the lung segmentation model is trained based on the lung sample image and its corresponding lung segmentation label.

[0027] Optionally, the medical image detection device further includes:

[0028] The display module is used to differentially display the bronchial segments with columnar signs in the lung medical image when the detection result shows that there are columnar signs.

[0029] Optionally, the medical image detection device further includes:

[0030] The storage module is used to associate and store the center point and inner diameter of the bronchial segment with the columnar sign when the detection result shows that the columnar sign exists.

[0031] Optionally, the acquisition module is specifically used to acquire the medical image to be detected; the medical image detection device also includes: a preprocessing module, used to preprocess the medical image to be detected; a segmentation module, used to perform lung segmentation processing on the preprocessed medical image to be detected to obtain a lung medical image; wherein the preprocessing includes at least one of the following: noise reduction processing, normalization processing, size correction processing, CT value truncation processing, and resampling processing.

[0032] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned medical image detection method is implemented.

[0033] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned medical image detection method when executing the computer program.

[0034] Through the above technical solution, the clinical criteria for the identification of columnar signs are taken into account, such as the diameter of the bronchus is larger than the diameter of the accompanying pulmonary artery, or the diameter of the bronchus does not tend to shrink as the bronchial tree runs. The image information of the bronchus and the accompanying arterial vascular information are used as the basis for the identification of columnar signs, which realizes the automatic detection and graded evaluation of bronchial columnar signs, reduces the difference in the subjective judgment of doctors, improves the detection accuracy of columnar signs, and then assists doctors in image judgment and identification, and improves the ability to accurately diagnose and treat bronchiectasis.

[0035] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 One of the flow diagrams of the medical image detection method provided in the embodiment of the present application is shown;

[0038] Figure 2 The second flowchart of the medical image detection method provided in the embodiment of the present application is shown;

[0039] Figure 3 A structural block diagram of a medical image detection device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0040] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0041] Typical symptoms of bronchiectasis include chronic cough, coughing up a lot of purulent sputum and repeated hemoptysis. In the early stage of bronchiectasis, there may be no obvious signs. Moist rales of varying sizes can be heard in the affected area. After the bronchiectasis gradually worsens, hypoxia symptoms appear, and clubbing of fingers and toes and poor overall nutrition can be seen. As described in the background technology, the bronchiectasis evaluation system currently used cannot make accurate judgments on columnar signs. In clinical applications, experienced doctors are often still required to evaluate computed tomography (CT) images, which is relatively subjective. In addition, the difference between the lesion area and the normal area on the conventional CT image and the boundary of the blood supply area are difficult to observe directly with the naked eye, and a high diagnostic accuracy rate cannot be achieved. Therefore, it is necessary to develop a system that can automatically evaluate CT images to achieve automated detection and graded evaluation of columnar signs of bronchiectasis.

[0042] In order to reduce the subjective differences among doctors, improve the efficiency and accuracy of columnar bronchiectasis diagnosis, reduce the waiting time of patients, and improve the treatment effect, the present application provides a medical imaging detection method, device, storage medium, and computer equipment. The following is a detailed description through embodiments.

[0043] like Figure 1 As shown, in this embodiment, a medical image detection method is provided, and the method includes:

[0044] Step 101, obtaining a lung medical image;

[0045] The patient's chest may be scanned by a chest photography instrument to obtain a lung medical image of the patient, and the lung medical image may be a CT image.

[0046] Specifically, the step of obtaining the lung medical image specifically includes: obtaining the medical image to be detected; preprocessing the medical image to be detected; and performing lung segmentation processing on the preprocessed medical image to be detected to obtain the lung medical image.

[0047] The preprocessing includes at least one of the following: noise reduction processing, normalization processing, size correction processing, CT value truncation processing, and resampling processing.

[0048] In this embodiment, the chest area of ​​the patient is scanned by a CT device to obtain a medical image to be detected. After obtaining the medical image to be detected, the medical image to be detected is preprocessed so that subsequent processing can obtain image inputs of uniform specifications, reduce interference from non-critical information, and thus ensure imaging quality, improve detection accuracy, stability and consistency. After preprocessing, the medical image to be detected is subjected to lung segmentation processing based on the lung segmentation model to obtain at least one lung region of interest corresponding to the medical image to be detected, that is, a lung medical image, so that the substantial area of ​​the lung can be segmented from the medical image to be detected to remove the image of the sternum, thereby avoiding interference of the sternum image during subsequent image registration, and facilitating accurate identification of bronchial columnar signs in the lungs.

[0049] In actual application scenarios, taking a three-dimensional plain scan CT image as an original input image as an example, during preprocessing, the size of the plain scan CT image is adjusted to a preset size through size correction processing, so as to unify all medical images and facilitate the subsequent comparison of the extracted features with the disease features. Then, linear or nonlinear smoothing and filtering techniques are used to reduce the noise of the plain scan CT image. For example, at least one of the methods such as Gaussian smoothing filtering and median filtering is used to remove noise (such as salt and pepper noise) in the image. Taking median filtering as an example, the gray value of each pixel in the plain scan CT image is set to the median of the gray values ​​of all pixels in a certain neighborhood window of the point, so that the surrounding pixel values ​​are close to the true value, thereby eliminating isolated noise points. After denoising, the CT value of the plain scan CT image is truncated and normalized between -1000 and 1000, and the obtained image value is distributed between 0 and 1, and it is re-interpolated in the three-dimensional direction at intervals of 0.6 mm in pixel size to obtain a standardized CT image with a width, height and depth of W×H×D. Finally, the lung region of interest is extracted using the lung segmentation model or the region growing method. The lung segmentation model can use the UNet network architecture or a related network architecture. Since the neural network model has the advantages of fast processing speed and high processing accuracy, lung segmentation using the lung segmentation model can effectively improve the rate and accuracy of lung segmentation, thereby improving the rate and accuracy of subsequent bronchial columnar sign detection, and has high robustness.

[0050] Step 102, extracting arteries and bronchi in the lung medical image;

[0051] The bronchi include multiple bronchial segments of different levels. The human bronchi have about 24 levels of branches to the alveoli, forming a bronchial tree. The bronchial segments of different levels can be recorded as L1, L2...L n It is understandable that a bronchial segment at the same level may include multiple bronchial branches. Taking a first-level bronchial segment as an example, multiple bronchial branches at the same level may be recorded as L 1-1 , L 1-2...L 1-m Using the centerline extraction method, the centerlines of bronchial segments at different levels are extracted, which can be recorded as line0 n-m , and each center line includes a starting point P0, an end point P w .

[0052] In actual application scenarios, three bronchial grading algorithms, namely traditional algorithm, deep learning model, template matching method and semantic segmentation algorithm, can be used to grade the bronchi.

[0053] Specifically, the lung medical image is input into the bronchial segmentation model to obtain the bronchi in the lung medical image, and the lung segmentation model is trained based on the lung sample image and its corresponding lung segmentation label. The lung medical image is input into the arterial segmentation model to obtain the arterial vessels in the lung medical image, and the arterial segmentation model is trained based on the lung sample image and its corresponding arterial segmentation label.

[0054] It is worth mentioning that since arteries and veins intersect each other in the image, they are adjacent and have very close grayscales, especially in non-contrast enhanced CT images, the difference between arteriovenous bifurcation and intersection is very small. Therefore, the CT value corresponding to each pixel in the lung medical image is first obtained; all target pixels corresponding to the CT values ​​within the threshold range are screened out, and the pulmonary vascular area and impurity area of ​​the lung medical image are roughly segmented, so as to combine the CT values ​​corresponding to each pixel in the lung CT image to ensure that all pulmonary vascular areas in the lung CT image can be extracted to avoid omissions during segmentation and extraction. Furthermore, according to the boundary formed by any pixel in the pulmonary vascular area and the impurity area and the pixel points in its neighborhood, the pulmonary vascular area is finely segmented. The image of the pulmonary vascular area is input into the arterial vascular segmentation model to effectively segment the arterial area and the venous area, thereby improving the accuracy and efficiency of pulmonary arteriovenous separation.

[0055] Step 103, determining a target arterial blood vessel segment corresponding to the bronchial segment in the lung medical image according to the distance between the centerline of the bronchial segment and the centerline of the arterial blood vessel;

[0056] The target arterial vessel segment is the arterial vessel that is closest to the bronchial segment, also called the accompanying arterial vessel.

[0057] Step 104, performing detection processing on the bronchial segment and the target arterial blood vessel segment to determine the detection result of the bronchial segment.

[0058] In this embodiment, the clinical criteria for distinguishing columnar signs are taken into account, such as the diameter of the bronchus being larger than the diameter of the accompanying pulmonary artery, or the diameter of the bronchus having no tendency to shrink as the bronchial tree runs. The arteries and bronchi in the lung medical images are extracted to obtain accurate bronchial and arterial information. The target arterial segment closest to the bronchial segment is determined by the distance between the centerline of the bronchial segment and the centerline of the arterial vessel. The image information of the bronchi and the target arterial vessel information are then used as the basis for distinguishing columnar signs, thereby realizing automatic detection and graded evaluation of bronchial columnar signs, reducing differences in subjective judgments of doctors, and improving the detection accuracy of columnar signs, thereby assisting doctors in image judgment and identification, and improving the ability to accurately diagnose and treat bronchiectasis.

[0059] In the embodiment of the present application, further, the step of performing detection processing on the bronchial segment and the target arterial segment to determine the detection result of the bronchial segment specifically includes:

[0060] Step 104-1, performing interception processing on the second bronchial segment in the lung medical image, at least one of the target arterial blood vessel segments corresponding to the first bronchial segment, and the first bronchial segment according to a preset interval to obtain a first image;

[0061] Among them, the first bronchial segment is the bronchus of the level at which the system user needs to detect columnar signs, and the second bronchial segment is the bronchial segment of the previous level of the first bronchial segment. For example, if the first bronchial segment is a level 4 bronchial segment, then the second bronchial segment is a level 3 bronchial segment. In other words, the first bronchial segment is a branch of the second bronchial segment.

[0062] Further, the first image includes at least one 3D image block of the first bronchial segment, and at least one 3D image block of the second bronchial segment and / or the target arterial segment, for example, the first image includes 2 partial images of the first bronchial segment and 2 partial images of the target arterial segment, or 2 partial images of the first bronchial segment and 2 partial images of the second bronchial segment, or 2 partial images of the first bronchial segment, the second bronchial segment and the target arterial segment. Moreover, on the bronchial segment or the target arterial segment, the distance between the center points of two adjacent 3D image blocks is a preset distance.

[0063] Step 104 - 2 , inputting the first image and / or the mask of the first image into a columnar sign detection model to determine a detection result of the first bronchial segment.

[0064] The columnar sign detection model is trained based on the lung sample images and the columnar sign labels corresponding to the lung sample images. The columnar sign labels are used to distinguish whether the bronchial segment samples in the lung sample images have columnar signs.

[0065] In this embodiment, the first image and / or the mask of the first image is used as input data, and the difference in diameter between the first bronchial segment and its accompanying pulmonary artery, or the diameter change trend of the first bronchial segment relative to the bronchial segment of the previous level is analyzed through the columnar sign detection model, so as to determine whether the first bronchial segment has columnar signs. The classification model constructed by deep learning technology is used to automatically detect bronchial columnar signs. At the same time, since the bronchi are graded in advance, the bronchi of the required level can be detected in a targeted manner, thereby realizing the graded evaluation of the bronchi. This reduces the difference in the subjective judgment of doctors, improves the diagnostic accuracy of columnar signs, and assists doctors in image diagnosis and identification, supports subsequent treatment and rehabilitation, and improves the ability to accurately diagnose and treat bronchiectasis.

[0066] It should be noted that since there is no second bronchial segment in the first-level bronchial segment, it is impossible to analyze whether the first-level bronchial segment has columnar signs through the trend of bronchial diameter changes.

[0067] It is understandable that in order to avoid the detection error of the model, the first image related to the first bronchial segment of a level can be input according to the bronchial level to obtain the detection result of the first bronchial segment of the level. Then the first image related to the first bronchial segment of another level is input. This cycle is repeated until all the first bronchial segments of the level required by the system user are detected, and the detection results of all the first bronchial segments are summarized.

[0068] Furthermore, the step of obtaining a columnar sign detection model includes: obtaining a lung sample image, the lung sample image including a plurality of bronchial segment samples of different levels and a plurality of target arterial segment samples corresponding to the bronchial segment samples of different levels; generating a columnar sign label for the first bronchial segment sample when the diameter of the first bronchial segment sample is greater than the diameter of the corresponding target arterial segment sample, or when the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is greater than a preset difference, the columnar sign label is used to mark the presence of a columnar sign in the bronchus; intercepting and processing the second bronchial segment sample marked with the columnar sign label, at least one of the target arterial segment sample corresponding to the first bronchial segment sample, and the first bronchial segment sample to obtain a second image; training a neural network model based on the second image and / or the mask of the second image to obtain a columnar sign detection model.

[0069] Among them, the second bronchial segment sample is the bronchial segment of the previous level of the first bronchial segment sample. In the case of human health, as the bronchial tree runs, the diameter of the bronchi tends to shrink, that is, the diameter of the first bronchial segment sample is smaller than that of the second bronchial segment sample. Then, in order to determine whether the first bronchial segment sample has columnar signs, the preset difference can be set to a value less than 0, and considering that the higher the level, the thinner the bronchial diameter, the smaller the difference between the bronchial diameters of two adjacent levels, therefore, the higher the level of the first bronchial segment sample, the larger the preset difference.

[0070] In this embodiment, the diameters of the first bronchial segment sample, the second bronchial segment sample, and the target arterial segment sample in the lung sample image are identified. The diameters of the first bronchial segment sample and the second bronchial segment sample in the lung sample image are compared. If the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is greater than the preset difference, it indicates that the change trend of the diameter of the bronchial segment meets the columnar sign; or the diameters of the first bronchial segment sample and its corresponding target arterial segment sample in the lung sample image are compared. If the diameter of the first bronchial segment sample is greater than the diameter of the corresponding target arterial segment sample, it indicates that there is a columnar sign. After the columnar sign is detected, a columnar sign label is generated to mark the first bronchial segment sample. The neural network model is trained using the first bronchial segment sample marked with the columnar sign label, and the second bronchial segment sample and / or the second image of the target arterial segment sample corresponding to the first bronchial segment sample, or the neural network model is trained using the mask of the second image to obtain a columnar sign detection model. Therefore, through deep learning and big data training, a feature model (columnar sign detection model) for detecting normal bronchi and columnar bronchi is established. The model is then used to distinguish normal bronchi from columnar dilated bronchi in clinical lung imaging data, thereby reducing the workload of doctors.

[0071] Specifically, the neural network (NN) model can be a convolutional neural network (CNN) model, a deep residual shrinkage network (DRSN) model, a fully connected neural network (FCNN) model, a recurrent neural network (RNN) model or a long short-term memory network (LSTM) model. The embodiments of the present application are not specifically limited.

[0072] It is understandable that if a columnar sign label has been added to the lung sample image, there is no need to perform the diameter comparison step again, and the neural network model can be trained directly based on the second image marked with the columnar sign label and / or the mask of the second image. For example, the patient's historical lung CT image is used as the lung sample image, and during the patient's medical treatment, the doctor's annotation result is used as the columnar sign label.

[0073] In actual application scenarios, in order to ensure the detection accuracy of columnar signs, the training samples required for the training model are as consistent as possible with the data input during detection. For example, when training a columnar sign detection model, the second image of the first bronchial segment sample and the second bronchial segment sample are used as training samples. When subsequently detecting through the columnar sign detection model, the first image including the first bronchial segment and the second bronchial segment needs to be input into the columnar sign detection model. For another example, when training a columnar sign detection model, the second image mask of the first bronchial segment sample and its corresponding target arterial segment sample is used as a training sample. When subsequently detecting through the columnar sign detection model, the first image mask including the first bronchial segment and the target arterial segment needs to be input into the columnar sign detection model.

[0074] In the embodiment of the present application, further, the step of determining the target arterial blood vessel segment corresponding to the bronchial segment in the lung medical image according to the distance between the center line of the bronchial segment and the center line of the arterial blood vessel specifically includes:

[0075] Step 103-1, determining a first sampling point on the center line of the bronchial segment;

[0076] The position of the first sampling point on the center line of the bronchial segment can be set as needed, the number of first sampling points on the center line of a segment can be multiple, and the spacing between multiple first sampling points can be the same or different. For example, it is preset to select 6 first sampling points on the center line of the bronchial segment, and the selection rule is that the starting point of the center line is the first first sampling point, the end point is the sixth first sampling point, and the first sampling points are equally spaced.

[0077] Step 103-2, respectively calculating the distances between the first sampling point and a plurality of second sampling points on the center line of the arterial blood vessel;

[0078] Step 103-3, determining the second sampling point corresponding to the minimum distance as the companion point corresponding to the first sampling point;

[0079] In this embodiment, for a first sampling point, the distances between the first sampling point and a plurality of second sampling points on the center line of the artery are calculated respectively to obtain a distance set related to the first sampling point. The second sampling point corresponding to the minimum value in the distance set, that is, the second sampling point on the artery closest to the first sampling point, is used as the companion point corresponding to the first sampling point, so as to facilitate the subsequent determination of the target artery segment corresponding to each bronchial segment.

[0080] Step 103-4, determining at least one arterial blood vessel segment corresponding to the bronchial segment according to the starting point coordinates and the ending point coordinates of the bronchial segment;

[0081] Step 103 - 5 , selecting a target arterial blood vessel segment from at least one arterial blood vessel segment according to the position of the companion point.

[0082] In this embodiment, the artery is intercepted according to the starting point coordinates and the end point coordinates of the bronchial segment to determine at least one artery segment that meets the length range of the bronchial segment. Considering that the multiple accompanying points corresponding to the multiple first sampling points of each bronchial segment may not be on the center line of the same artery, the artery segment with the most accompanying points is selected as the target artery segment closest to the bronchial segment in a statistical manner. This facilitates the detection of columnar signs through the relevant information of the target artery segment and the bronchial segment.

[0083] In an embodiment of the present application, further, the medical image detection method also includes: when the detection result shows that there is a columnar sign, differentially displaying the bronchial segment with the columnar sign in the lung medical image.

[0084] In this embodiment, if it is determined that the bronchial segment of the lung medical image has columnar signs, the bronchial segment with columnar signs and the bronchial segment without columnar signs in the lung medical image are displayed differently, so that the system user (doctor) can intuitively know the dilation of the bronchial segment, which is conducive to improving the efficiency of identifying bronchiectasis.

[0085] Specifically, the difference display method can be to generate a columnar sign label at a specified position of the bronchial segment in the lung medical image; or to highlight the label information of the bronchial segment, for example, to highlight the name of the bronchial segment in the lung medical image; or to add text prompt information that the bronchial segment has a columnar sign in the lung medical image. The embodiment of the present application is not specifically limited.

[0086] Furthermore, the step of differentially displaying the bronchial segments with columnar signs in the lung medical image specifically includes: extracting the sign feature information of the bronchial segments with columnar signs; determining the display level according to the sign feature information; and differentially displaying the bronchial segments with columnar signs according to the display mode corresponding to the display level. For example, the severity of columnar bronchiectasis is determined by the sign feature information, and the display mode of the bronchial segment is determined by the corresponding relationship between the severity and the display level. For example, if the diameter difference between the first bronchial segment and the second bronchial segment is greater than the preset difference associated with the first display level, the first bronchial segment is determined to be the first display level, and the display mode corresponding to the first display level is to highlight the marking information of the bronchial segment in red; if the diameter difference between the first bronchial segment and the second bronchial segment is greater than the preset difference associated with the second display level, the first bronchial segment is determined to be the second display level, and the display mode corresponding to the first display level is to highlight the marking information of the bronchial segment in yellow, wherein the preset difference associated with the first display level is greater than the preset difference associated with the second display level, that is, the bronchial segment of the first display level is more severe than the bronchial segment of the second display level. Thus, through different display modes, the system user can intuitively distinguish bronchial segments with different bronchiectasis through lung medical images, greatly improving the efficiency of bronchiectasis identification.

[0087] In an embodiment of the present application, further, the medical image detection method also includes: when the detection result is that there is a columnar sign, the center point and inner diameter of the bronchial segment with the columnar sign are associated and stored.

[0088] In this embodiment, if it is determined that a bronchial segment of a lung medical image has a columnar sign, the center point and inner diameter of the bronchial segment with the columnar sign are associated and stored to facilitate information tracing of the columnar sign, so that the system user can intuitively know the expansion of the bronchial segment, thereby reducing the system user's workload of reading films and helping to improve the efficiency of bronchiectasis identification.

[0089] like Figure 2 As shown, in a specific embodiment of the present application, a medical image detection method is proposed, comprising:

[0090] Step 201, image acquisition and image preprocessing;

[0091] Specifically, the chest of the patient can be scanned by a CT device to obtain the chest image to be detected. The chest image to be detected is standardized, the CT value of the chest image to be detected is truncated and normalized between -1000 and 1000, the obtained image value is distributed between 0 and 1, and it is re-interpolated in the three-dimensional direction at intervals of 0.6 mm in pixel size to obtain a standardized chest image to be detected with a width, height and depth of W×H×D.

[0092] Step 202, lung region segmentation;

[0093] Specifically, the traditional region growing method or the model-based lung parenchyma segmentation method can be used to obtain the lung region segmentation results (lung medical images). If the latter is used, the lung parenchyma area needs to be labeled according to needs, and then the lung parenchyma segmentation model is obtained through neural network model training.

[0094] Step 203, bronchial segmentation;

[0095] Specifically, bronchial segmentation is performed in the lung segmentation area, and all bronchi are marked as mask0. Bronchial segmentation can use traditional region growing methods, such as selecting seed points from the thicker part of the trachea to start region growth based on thresholds; or a model-based method can be used to segment the bronchi. If a model-based method is used, the bronchi need to be manually marked according to needs, and then the bronchial segmentation model is obtained through neural network model training.

[0096] Step 204, segmenting the pulmonary artery;

[0097] Specifically, arterial segmentation is performed in the lung segmentation area, and all pulmonary arterial vessels are labeled as mask1. Since there is no difference between arteriovenous vessels in plain scan CT, the traditional pulmonary arterial segmentation method is not effective. A model-based method can be used to segment arterial vessels. Arterial vessels need to be manually labeled according to needs, and then the arterial segmentation model is obtained through neural network model training. However, since there is little difference between arteriovenous vessels, it is sufficient to separate the vessels.

[0098] Step 205, bronchial classification;

[0099] Specifically, according to the bronchial segmentation result mask0, the bronchial centerline line0 is extracted using the centerline extraction method. n , and divide the bronchi into different levels on mask0 according to the connected domain calculated by the center line, and obtain the bronchial grades L1, L2...L n .

[0100] Step 206, searching for bronchial arteries;

[0101] Specifically, the distance field from all points in the lung parenchyma area to the bronchial centerline is calculated to obtain a map; the arterial centerline line1 is extracted from the pulmonary artery segmentation result mask1; and different levels of bronchial segments line1 are extracted from the bronchial classification. n Each bronchial segment centerline includes the starting point P0, the end point P w And the first sampling points P1, P2, ... P (w-1), search the map for the accompanying point on line1 with the shortest distance from the sampling point of each bronchial segment to line1, denoted as A0, A1, ... A (w-1) ; Sometimes point A0, A1...A (w-1) The two segments may not be on the same arterial centerline. The centerline with the most votes can be selected as the accompanying arterial segment (target arterial segment) centerline line1 of the bronchial segment by multi-point voting. n , so that the closest accompanying arterial vessel segment can be searched on mask1 according to the different bronchial segments on mask0.

[0102] Step 207, bronchial classification;

[0103] Specifically, the model-based bronchial classification method is based on the bronchial segment line0 in the lung sample image. n and the accompanying arterial segment line1 n , get the following information:

[0104] (1) According to the bronchial segment line0 n The center line is straightened to obtain the original image of the bronchus segment, and three-dimensional small block images are captured at equal intervals. n ;

[0105] (2) Use the same method to cut the bronchial segment line 0 n 3D patch mask M0 on mask0 n ;

[0106] (3) In this bronchial segment line0 n The superior bronchial segment line0 (n-1) The same operation is performed to extract the corresponding three-dimensional image I0 (n-1) , 3D mask M0 (n-1) ;

[0107] (4) In the bronchial segment line0 n The accompanying arterial segment line1 n The same operation is performed to extract the corresponding three-dimensional image I1 n , 3D mask M1 n ;

[0108] The collected image block I0 n 、M0 n , I0 (n-1) 、M0 (n-1) 、I1 n 、M1 nThe data are combined as the input of the neural network, and the three-dimensional neural network model (columnar sign detection model) is trained based on the doctor's labeled results (0 for normal bronchus, 1 for dilated bronchus). There are no special requirements for the model structure and parameter design.

[0109] In the process of detecting columnar signs, the same method is used to obtain the line0 of the bronchial segment in the lung medical image. n and the accompanying arterial segment line1 n Related I0 n 、M0 n , I0 (n-1) 、M0 (n-1) 、I1 n 、M1 n , and input the columnar sign detection model. According to the bronchial segment information in mask0 and the accompanying arterial blood vessel information in mask1, the deep learning model is used to distinguish whether the bronchus has columnar dilatation.

[0110] Step 208, bronchial information output;

[0111] Specifically, the bronchial segments marked as 1 that have been classified in step 207 are marked with different values ​​on mask0, and the center point of the bronchial segment is recorded and the inner diameter of the bronchus is calculated as output, so that the measurement and statistical information of each dilation is calculated and output based on the dilated bronchus and its position information.

[0112] In this embodiment, a feature model of the bronchus is established by using big data training through deep learning, and bronchial image information, accompanying arterial vascular information, and superior bronchial image information are used as the basis for network discrimination. It can not only distinguish normal bronchi from columnar dilated bronchi in clinical lung CT image data, but also improve the diagnostic accuracy of columnar signs and reduce the workload of doctors.

[0113] Furthermore, as a specific implementation of the above-mentioned medical image detection method, the present application embodiment provides a medical image detection device, such as Figure 3 As shown, the medical image detection device includes: an acquisition module, an extraction module, a determination module and a detection module.

[0114] Among them, the acquisition module is used to acquire lung medical images; the extraction module is used to extract arteries and bronchi in the lung medical images, and the bronchi include multiple bronchial segments of different levels; the determination module is used to determine the target arterial segment corresponding to the bronchial segment in the lung medical image based on the distance between the center line of the bronchial segment and the center line of the arterial vessel; the detection module is used to detect and process the bronchial segment and the target arterial segment to determine the detection result of the bronchial segment.

[0115] Furthermore, the medical image detection device also includes: a capture module, which captures the second bronchial segment, at least one of the target arterial blood vessel segments corresponding to the first bronchial segment, and the first bronchial segment in the lung medical image according to a preset interval to obtain a first image, wherein the second bronchial segment is a bronchial segment of the previous level of the first bronchial segment; the detection module is specifically used to input the first image and / or the mask of the first image into a columnar sign detection model to determine the detection result of the first bronchial segment; wherein the columnar sign detection model is trained based on the lung sample image and the columnar sign label corresponding to the lung sample image.

[0116] Furthermore, the acquisition module is also used to acquire a lung sample image, which includes multiple bronchial segment samples of different levels and target arterial segment samples corresponding to multiple bronchial segment samples of different levels; the medical image detection device also includes: a marking module, which is used to generate a columnar sign label for the first bronchial segment sample when the diameter of the first bronchial segment sample is larger than the diameter of the corresponding target arterial segment sample, or the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is larger than a preset difference, and the columnar sign label is used to mark the presence of a columnar sign in the bronchus; the interception module is also used to intercept the second bronchial segment sample marked with the columnar sign label, at least one of the target arterial segment sample corresponding to the first bronchial segment sample, and the first bronchial segment sample to obtain a second image; the medical image detection device also includes: a training module, which is used to train a neural network model according to the second image and / or the mask of the second image to obtain a columnar sign detection model.

[0117] Furthermore, the determination module is specifically used to determine a first sampling point on the center line of the bronchial segment; respectively calculate the distances between the first sampling point and multiple second sampling points on the center line of the arterial blood vessel; the determination module is specifically used to determine the second sampling point corresponding to the smallest distance as the accompanying point corresponding to the first sampling point; determine at least one arterial blood vessel segment corresponding to the bronchial segment according to the starting point coordinates and the end point coordinates of the bronchial segment; and select a target arterial blood vessel segment from at least one arterial blood vessel segment according to the position of the accompanying point.

[0118] Furthermore, the extraction module is specifically used to input the lung medical image into the arterial segmentation model to obtain the arteries in the lung medical image; input the lung medical image into the bronchial segmentation model to obtain the bronchi in the lung medical image; wherein the arterial segmentation model is trained based on the lung sample image and its corresponding arterial segmentation label, and the lung segmentation model is trained based on the lung sample image and its corresponding lung segmentation label.

[0119] Furthermore, the medical image detection device also includes: a display module for differentially displaying the bronchial segments with columnar signs in the lung medical image when the detection result shows that there are columnar signs.

[0120] Furthermore, the medical image detection device also includes: a storage module for associatively storing the center point and inner diameter of the bronchial segment with the columnar sign when the detection result shows that there is a columnar sign.

[0121] Furthermore, the acquisition module is specifically used to acquire the medical image to be detected; the medical image detection device also includes: a preprocessing module, used to preprocess the medical image to be detected; a segmentation module, used to perform lung segmentation processing on the preprocessed medical image to be detected to obtain a lung medical image; wherein the preprocessing includes at least one of the following: noise reduction processing, normalization processing, size correction processing, CT value truncation processing, and resampling processing.

[0122] In this embodiment, each module of the medical image detection device implements the steps of the medical image detection method in any embodiment of the first aspect when performing its respective function. Therefore, the medical image detection device also includes all the beneficial effects of the medical image detection method in any embodiment of the first aspect, which will not be repeated here.

[0123] It should be noted that for other corresponding descriptions of the functional modules involved in the medical image detection device provided in the embodiment of the present application, reference can be made to Figure 1 and Figure 2 The corresponding description will not be repeated here.

[0124] In a specific application scenario, the medical image detection device in the embodiment of the present application can be a device, or a component, an integrated circuit, or a chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, an intelligent camera device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (NAS), or a personal computer (PC), etc., which is not specifically limited in the embodiment of the present application.

[0125] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned medical image detection method is implemented.

[0126] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned medical image detection method when executing the computer program.

[0127] Based on the above Figure 1 and Figure 2 The method shown, accordingly, an embodiment of the present application also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned medical image detection method is implemented.

[0128] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product or a hardware product or a combination of software and hardware. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of the present application.

[0129] Based on the above Figure 1 The detection method of the medical image shown, and Figure 3 The embodiment of the medical image detection device shown in the figure, in order to achieve the above-mentioned purpose, the embodiment of the present application also provides a computer device, which can be specifically a personal computer, a server, a network device, etc., and the computer device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to implement the above-mentioned medical image detection method.

[0130] Optionally, the computer device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.

[0131] Those skilled in the art will appreciate that the computer device structure provided in this embodiment does not limit the computer device, and may include more or fewer components, or a combination of certain components, or different component arrangements.

[0132] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages and saves the hardware and software resources of the computer device, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to realize the communication between the various controls inside the storage medium, and the communication with other hardware and software in the physical device.

[0133] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform, or by hardware.

[0134] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the units or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the units in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The units of the above-mentioned implementation scenarios can be combined into one unit, or can be further split into multiple sub-units.

[0135] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A method for detecting medical images, characterized in that: The method comprises: Obtain medical images of the lungs; Extracting arteries and bronchi from the lung medical image, wherein the bronchi include multiple bronchial segments of different levels; Determining a target arterial blood vessel segment corresponding to the bronchial segment in the lung medical image according to a distance between a centerline of the bronchial segment and a centerline of the arterial blood vessel; According to a preset interval, the second bronchial segment in the lung medical image, the target arterial segment corresponding to the first bronchial segment, and the first bronchial segment are intercepted and processed to obtain a first image, wherein the second bronchial segment is a bronchial segment at a previous level of the first bronchial segment; Inputting the first image and / or the mask of the first image into a columnar sign detection model to determine a detection result of whether the first bronchial segment has a columnar sign; The columnar sign detection model is trained based on a lung sample image and a columnar sign label corresponding to the lung sample image, and the columnar sign label is generated when the diameter of the first bronchial segment sample in the lung sample image is greater than the diameter of the corresponding target arterial segment sample, or when the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is greater than a preset difference; Determining the severity of columnar bronchiectasis according to the sign feature information of the bronchial segment with columnar signs in the lung medical image, and determining the display mode according to the corresponding relationship between the severity and the display level; The bronchial segments with columnar signs are differentially displayed according to the display mode.

2. The medical image detection method according to claim 1, characterized in that: The method further comprises: Acquire the lung sample image, wherein the lung sample image includes a plurality of bronchial segment samples at different levels and target arterial blood vessel segment samples corresponding to the plurality of bronchial segment samples at different levels; When the diameter of the first bronchial segment sample is greater than the diameter of the corresponding target arterial segment sample, or the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is greater than a preset difference, a columnar sign label of the first bronchial segment sample is generated, and the columnar sign label is used to mark the presence of a columnar sign in the bronchus; performing interception processing on the second bronchial segment sample marked with the columnar sign label, at least one of the target arterial segment sample corresponding to the first bronchial segment sample, and the first bronchial segment sample to obtain a second image; A neural network model is trained according to the second image and / or the mask of the second image to obtain the columnar sign detection model.

3. The medical image detection method according to claim 1, characterized in that: The step of determining the target arterial blood vessel segment corresponding to the bronchial segment in the lung medical image according to the distance between the centerline of the bronchial segment and the centerline of the arterial blood vessel specifically includes: determining a first sampling point on the centerline of the bronchial segment; respectively calculating the distances between the first sampling point and a plurality of second sampling points on the center line of the arterial blood vessel; Determine the second sampling point corresponding to the smallest distance as the companion point corresponding to the first sampling point; Determine at least one arterial blood vessel segment corresponding to the bronchial segment according to the starting point coordinates and the ending point coordinates of the bronchial segment; A target arterial vessel segment is selected from the at least one arterial vessel segment according to the position of the companion point.

4. The medical image detection method according to any one of claims 1 to 3, characterized in that: The extracting of arteries and bronchi in the lung medical image specifically includes: Inputting the lung medical image into an arterial blood vessel segmentation model to obtain the arterial blood vessels in the lung medical image; Inputting the lung medical image into a bronchial segmentation model to obtain bronchi in the lung medical image; The arterial segmentation model is trained based on lung sample images and their corresponding arterial segmentation labels, and the bronchial segmentation model is trained based on lung sample images and their corresponding lung bronchial segmentation labels.

5. The medical image detection method according to any one of claims 1 to 3, characterized in that: The method further comprises: When the detection result indicates that a columnar sign exists, the center point and the inner diameter of the bronchial segment where the columnar sign exists are associated and stored.

6. The medical image detection method according to any one of claims 1 to 3, characterized in that: The obtaining of lung medical images specifically includes: Acquire the medical image to be tested; Preprocessing the medical image to be detected; Performing lung segmentation processing on the preprocessed medical image to be detected to obtain the lung medical image; The preprocessing includes at least one of the following: noise reduction processing, normalization processing, size correction processing, CT value truncation processing, and resampling processing.

7. A medical image detection device, characterized in that: The device comprises: An acquisition module, used for acquiring lung medical images; An extraction module, used for extracting arteries and bronchi in the lung medical image, wherein the bronchi include multiple bronchial segments of different levels; a determination module, configured to determine a target arterial blood vessel segment corresponding to the bronchial segment in the lung medical image according to a distance between a centerline of the bronchial segment and a centerline of the arterial blood vessel; A detection module, used for performing detection processing on the bronchial segment and the target arterial segment, and determining a detection result of the bronchial segment; a capture module, configured to capture the second bronchial segment in the pulmonary medical image, the target arterial segment corresponding to the first bronchial segment, and the first bronchial segment according to a preset interval to obtain a first image, wherein the second bronchial segment is a bronchial segment at a previous level of the first bronchial segment; The detection module is specifically used to input the first image and / or the mask of the first image into a columnar sign detection model to determine the detection result of whether the first bronchial segment has a columnar sign; wherein the columnar sign detection model is trained based on a lung sample image and a columnar sign label corresponding to the lung sample image, and the columnar sign label is generated when the diameter of the first bronchial segment sample in the lung sample image is greater than the diameter of the corresponding target arterial segment sample, or the difference between the diameter of the first bronchial segment sample and the diameter of the second bronchial segment sample is greater than a preset difference; The determination module is used to determine the severity of columnar bronchiectasis according to the sign feature information of the bronchial segment with columnar signs in the lung medical image, and determine the display mode according to the corresponding relationship between the severity and the display level; The display module is used to differentially display the bronchial segment with columnar signs according to the display mode.

8. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the medical image detection method according to any one of claims 1 to 6 is implemented.

9. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the medical image detection method according to any one of claims 1 to 6 is implemented.

Images