Image detection method and related device, electronic device and storage medium

Abstract

The present application discloses an image detection method and related devices, electronic equipment and storage medium, wherein the image detection method comprises: detecting a medical image to be detected, obtaining a detection result of a bleeding focus in the medical image to be detected; wherein the detection result comprises the amount of bleeding of the bleeding focus; and displaying the medical image to be detected and the detection result on the same screen in an interactive interface. The above scheme can assist doctors in improving the speed of reading films.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an image detection method and related devices, electronic equipment and storage media. Background Art

[0002] Medical images such as CT (Computed Tomography) images are extremely important in scenarios such as medical planning. For example, in the process of bleeding detection, doctors can obtain relevant auxiliary information by reading medical images to better assist doctors.

[0003] However, when facing scenarios such as hemorrhagic stroke, since the speed of reading films is often an extremely important part of the treatment process, providing important assistance for subsequent surgical planning, in view of this, how to assist doctors in improving the speed of reading films has become an urgent problem to be solved. Summary of the invention

[0004] The present application provides an image detection method and related devices, electronic equipment and storage medium.

[0005] The first aspect of the present application provides an image detection method, comprising: detecting a medical image to be tested to obtain a detection result of a bleeding focus in the medical image to be tested; wherein the detection result includes the amount of bleeding in the bleeding focus; and displaying the medical image to be tested and the detection result on the same screen in an interactive interface.

[0006] Therefore, the medical image to be tested is tested to obtain the detection result of the bleeding focus in the medical image to be tested, and the detection result includes the amount of bleeding of the bleeding focus. On this basis, the medical image to be tested and the detection result are displayed on the same screen on the interactive interface. Therefore, the doctor can be saved from manually estimating the amount of bleeding and other related indicators by reading the medical image to be tested, thereby greatly reducing the doctor's manual operation time during the film reading process, and then assisting the doctor to improve the film reading speed.

[0007] Among them, the interactive interface includes a first display area and a second display area, and the medical image to be tested is a three-dimensional medical image, wherein the first display area is used to simultaneously display at least one two-dimensional medical image of the medical image to be tested on several preset planes, and the second display area is used to display a bleeding focus list, which includes data items belonging to each bleeding focus, and the data items include the amount of bleeding of the bleeding focus.

[0008] Therefore, the interactive interface is configured to include a first display area and a second display area, and the medical image to be tested is a three-dimensional medical image. The first display area is used to simultaneously display at least one two-dimensional medical image of the medical image to be tested on several preset planes, and the second display area is used to display a list of bleeding foci, and the list of bleeding foci is configured to include data items belonging to each bleeding foci, respectively, and the data items include the amount of bleeding of the bleeding foci. That is, by displaying the blood foci and their amount of bleeding detected in the medical image to be tested in the form of a list, the intuitiveness of the display of the test results can be improved, and by displaying the two-dimensional medical images on several preset planes, the medical images can be displayed from different angles of the planes, which can improve the richness of the image display. Therefore, on the basis of improving the intuitiveness of the display of the test results and the richness of the image display, it can be helpful to further assist doctors in improving the speed of reading films.

[0009] The detection result also includes the category of the bleeding focus, and the data item also includes the category of the bleeding focus.

[0010] Therefore, the detection results also include the category of bleeding foci, and on the basis of the data item including the amount of bleeding of the bleeding foci, it further includes the category of the bleeding foci, that is, for each bleeding foci, its category and amount of bleeding can be displayed simultaneously on the interactive interface, so the information richness of each bleeding foci can be improved, which is conducive to further assisting doctors to improve the speed of reading films.

[0011] The detection result also includes the outline of the bleeding focus, and the method also includes: in response to a data item selected by a user in the bleeding focus list, taking the outline of the bleeding focus to which the selected data item belongs as the target outline; and marking the target outline in a preset style on the two-dimensional medical image.

[0012] Therefore, the detection result also includes the outline of the bleeding focus, and in response to the data item selected by the user in the bleeding focus list, the outline of the bleeding focus to which the selected data item belongs is used as the target outline, and the target outline is marked in a preset style on the two-dimensional medical image. That is, it can respond to the doctor's operation to mark the outline that the doctor expects to see on the medical image, which is conducive to further assisting the doctor to improve the speed of reading the film.

[0013] Among them, the detection results also include the outline of the bleeding focus. After the medical image to be tested and the detection results are displayed on the same screen on the interactive interface, the method also includes: in response to the user's operation of modifying the outline on the interactive interface, re-analyzing the bleeding amount of the bleeding focus based on the modified outline, and updating the bleeding amount displayed on the interactive interface to the bleeding amount obtained by the re-analysis.

[0014] Therefore, the detection result also includes the outline of the bleeding focus, and after being displayed on the same screen, it further responds to the user's operation of modifying the outline in the interactive interface, re-analyzes the bleeding amount of the bleeding focus based on the modified outline, and updates the bleeding amount displayed on the interactive interface to the re-analyzed bleeding amount. That is, it can respond to the doctor's operation to modify the automatically detected outline on the medical image, and automatically re-analyze the bleeding amount based on this, which is beneficial to improving the doctor's autonomy in the film reading process and can save the doctor from manually performing complex analysis and calculations. Therefore, it can assist doctors in improving the speed of film reading while improving autonomy.

[0015] The medical image to be tested is a three-dimensional medical image, and the medical image to be tested includes several two-dimensional medical images on a preset plane, and the interactive interface displays a two-dimensional medical image on the preset plane. In response to the user's operation of modifying the contour on the interactive interface, the amount of bleeding of the bleeding focus is re-analyzed based on the modified contour, including: obtaining the modified contour in the two-dimensional medical image on the preset plane; in response to the user's switching instruction, the next two-dimensional medical image is displayed on the interactive interface, and the step of obtaining the modified contour in the two-dimensional medical image on the preset plane and subsequent steps are re-executed until the modified contours in each two-dimensional medical image on the preset plane are obtained; based on the modified contours in each two-dimensional medical image on the preset plane, the amount of bleeding is analyzed.

[0016] Therefore, the medical image to be tested is a three-dimensional medical image, which contains several two-dimensional medical images on a preset plane, and the interactive interface displays a two-dimensional medical image on the preset plane. On this basis, the modified contour in the two-dimensional medical image on the preset plane is obtained, and in response to the user's switching instruction, the next two-dimensional medical image is displayed on the interactive interface, and the above-mentioned steps of obtaining the modified contour and subsequent steps are re-executed until the modified contours in each two-dimensional medical image on the preset plane are obtained. On this basis, the amount of bleeding is analyzed based on the modified contours in each two-dimensional medical image on the preset plane. Therefore, it can support doctors to modify the contours on each two-dimensional medical image on the preset plane through image switching, which can reduce the complexity of contour modification operations and is conducive to further assisting doctors to improve the speed of reading films.

[0017] The detection result is obtained by using a detection model, and the detection model is trained based on sample medical images, and the sample medical images are at least marked with sample contours of bleeding foci.

[0018] Therefore, the detection result is obtained by using the detection model, and the detection model is trained based on sample medical images, and the sample medical images are at least marked with sample contours of bleeding foci, so it can be beneficial to improve the efficiency of image detection.

[0019] The medical image to be tested is obtained by scanning the brain, and the bleeding foci include at least one of the following: intraventricular hemorrhage, brain parenchymal hemorrhage, subarachnoid hemorrhage, epidural hemorrhage, and subdural hematoma.

[0020] Therefore, the medical image to be tested is obtained by scanning the brain, and the bleeding foci include at least one of the following: intraventricular hemorrhage, intracerebral parenchymal hemorrhage, subarachnoid hemorrhage, epidural hemorrhage, and subdural hematoma, that is, it can support image detection of different types of cerebral hemorrhage.

[0021] The detection result may further include the category of the bleeding focus; and / or the detection result may further include the outline of the bleeding focus; and / or the detection result may further include the pixel value of the bleeding focus.

[0022] Therefore, the detection result also includes at least one of the category and outline of the bleeding focus, and the pixel value of the bleeding focus, so it can be helpful to further enrich the auxiliary information that can be provided to doctors, and help further assist doctors to improve the speed of reading films.

[0023] The second aspect of the present application provides an image detection device, including: a detection module and a display module, the detection module is used to detect the medical image to be tested and obtain the detection result of the bleeding focus in the medical image to be tested; wherein the detection result includes the amount of bleeding of the bleeding focus; the display module is used to display the medical image to be tested and the detection result on the same screen in an interactive interface.

[0024] A third aspect of the present application provides an electronic device, comprising a memory and a processor coupled to each other, wherein the processor is used to execute program instructions stored in the memory to implement the image detection method in the first aspect.

[0025] A fourth aspect of the present application provides a computer-readable storage medium having program instructions stored thereon, which implement the image detection method in the first aspect when the program instructions are executed by a processor.

[0026] The above scheme detects the medical image to be tested and obtains the detection result of the bleeding focus in the medical image to be tested, and the detection result includes the amount of bleeding of the bleeding focus. On this basis, the medical image to be tested and the detection result are displayed on the same screen on the interactive interface. Therefore, it can save the doctor from manually estimating the amount of bleeding and other related indicators by reading the medical image to be tested, thereby greatly reducing the doctor's manual operation time during the film reading process, and then assisting the doctor to improve the film reading speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flowchart of an embodiment of the image detection method of the present application;

[0028] Figure 2 is a schematic diagram of an embodiment of an interactive interface;

[0029] Figure 3 is a schematic diagram of an embodiment of a second display area;

[0030] Figure 4 is a flow chart of another embodiment of the image detection method of the present application;

[0031] Figure 5 It is a schematic diagram of the framework of an embodiment of the image detection device of the present application;

[0032] Figure 6 It is a schematic diagram of the framework of an embodiment of the electronic device of the present application;

[0033] Figure 7 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium of the present application. DETAILED DESCRIPTION

[0034] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0035] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0036] The terms "system" and "network" are often used interchangeably in this article. The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship. In addition, "many" in this article means two or more than two.

[0037] See also Figure 1 , Figure 1 It is a flowchart of an embodiment of the image detection method of the present application.

[0038] Specifically, the following steps may be included:

[0039] Step S11: Detect the medical image to be detected to obtain the detection result of the bleeding focus in the medical image to be detected.

[0040] In the disclosed embodiment, the test result includes the amount of bleeding of the bleeding focus. Specifically, the amount of bleeding may include but is not limited to the volume of the bleeding focus, the mass of the bleeding focus, etc., which are not limited here. It should be noted that after obtaining the volume of the bleeding focus, the mass of the bleeding focus can be obtained by the product of the blood density and the volume of the bleeding focus.

[0041] In one implementation scenario, the medical image to be tested may include but is not limited to a CT image, etc., which is not limited here. In addition, the medical image to be tested may specifically be a three-dimensional medical image.

[0042] In an implementation scenario, the target detection can be performed on the medical image to be tested first to obtain the target area and category of the bleeding focus in the medical image to be tested. It should be noted that the target area can be a square area surrounding the bleeding focus. Taking the medical image to be tested as a three-dimensional medical image as an example, the target area can be a rectangular area, and the rectangular area surrounds the bleeding focus. Other cases can be deduced by analogy, and no examples are given here. On this basis, the image data of the target area can be further segmented to obtain the outline of the bleeding focus. It should be noted that, unlike the aforementioned target area, although the outline of the bleeding focus also surrounds the bleeding focus, the outline is the boundary of the bleeding focus, that is, the bleeding focus belongs to the bleeding focus within the outline of the bleeding focus, and the bleeding focus does not belong to the bleeding focus outside the outline of the bleeding focus. Further, still taking the medical image to be tested as a three-dimensional medical image as an example, the volume of the bleeding focus can be calculated based on the outline area and the physical distance between adjacent pixels as the bleeding amount of the bleeding focus. Of course, the volume can also be multiplied by the blood density to obtain the mass of the bleeding focus as the bleeding amount of the bleeding focus. It should be noted that the physical distance refers to the distance between adjacent pixels in a real scene. For example, the physical distance can be 3 mm, 5 mm, 10 mm, etc., and is not limited here.

[0043] In a specific implementation scenario, in order to improve the efficiency of target detection, a target detection network can be pre-trained, based on which the target detection network can be used to perform target detection on the medical image to be tested, and the target area and category of the hemorrhage in the medical image to be tested can be obtained. The target detection network can include but is not limited to: Faster-RCNN, YOLO, etc., and the network structure of the target detection network is not limited here. Specifically, sample medical images can be collected in advance, and the sample medical images can be annotated with sample areas and sample categories of hemorrhages. Based on this, the target detection network can be used to perform target detection on the sample medical images, and the predicted area and predicted category of the hemorrhage in the sample medical images can be obtained, so that loss functions such as cross entropy and dice loss can be used to measure the difference between the sample category and the predicted category and the difference between the sample area and the predicted area, respectively, to obtain the loss value of the target detection network, and then use optimization methods such as gradient descent to adjust the network parameters of the target detection network based on the loss value. It should be noted that the specific measurement process of the difference and the specific adjustment process of the parameters can refer to the technical details of loss functions such as cross entropy and dice loss, and the technical details of optimization methods such as gradient descent, which will not be repeated here.

[0044] In a specific implementation scenario, in order to improve the efficiency of target segmentation, a target segmentation network can be pre-trained, based on which the target segmentation network can be used to perform target segmentation on the image data of the target area to obtain the contour of the hemorrhage. The target segmentation network may include but is not limited to: U-Net, etc., and the network structure of the target segmentation network is not limited here. Specifically, a sample medical image can be collected in advance, and the sample medical image can be annotated with a sample contour of the hemorrhage focus. Based on this, the target segmentation network can be used to perform target segmentation on the sample medical image to obtain the predicted contour of the hemorrhage focus, so that the difference between the predicted contour and the sample contour can be measured by loss functions such as cross entropy and dice loss, and the loss value of the target segmentation network can be obtained, and then the network parameters of the target segmentation network can be adjusted based on the loss value by using optimization methods such as gradient descent. It should be noted that the specific measurement process of the difference and the specific adjustment process of the parameters can refer to the technical details of loss functions such as cross entropy and dice loss, and the technical details of optimization methods such as gradient descent, respectively, which will not be repeated here.

[0045] In an implementation scenario, the medical image to be tested can also be directly instantiated and segmented to obtain the contour and category of the bleeding focus in the medical image to be tested. Still taking the medical image to be tested as a three-dimensional medical image as an example, the volume of the bleeding focus can be calculated based on the contour and the physical distance between adjacent pixels as the bleeding amount of the bleeding focus. Of course, the volume can also be multiplied by the blood density to obtain the mass of the bleeding focus as the bleeding amount of the bleeding focus.

[0046] In a specific implementation scenario, in order to improve the efficiency of instantiation segmentation, an instantiation segmentation network can be pre-trained, based on which the instantiation segmentation network can be used to perform instantiation segmentation on the medical image to be tested, and the contour and category of the hemorrhage in the medical image to be tested can be obtained. The instantiation segmentation network may include but is not limited to: DeepMask, MaskRCNN, etc., and the network structure of the instantiation segmentation network is not limited here. Specifically, sample medical images can be collected in advance, and the sample medical images are annotated with sample contours and sample categories of hemorrhages. For example, each sample pixel in the sample medical image can be annotated with its sample pixel category, and the sample pixel category can be any of several hemorrhage categories and image backgrounds, that is, when the sample pixel category is annotated as a certain hemorrhage category, it can be considered that the sample pixel belongs to the hemorrhage, and the hemorrhage to which it belongs is this hemorrhage category, or, when the sample pixel category is annotated as image background, it can be considered that the sample pixel does not belong to the hemorrhage. On this basis, the instantiated segmentation network can be used to perform instantiated segmentation on the sample medical image to obtain the predicted pixel category of each sample pixel in the sample medical image, and the predicted pixel category is also any one of several bleeding focus categories and image background. Based on this, a loss function such as cross entropy can be used to measure the difference between the predicted pixel category and the sample pixel category of each sample pixel to obtain the loss value of the instantiated segmentation network, and an optimization method such as gradient descent can be used to adjust the network parameters of the instantiated segmentation network based on the loss value. It should be noted that the specific measurement process of the difference and the specific adjustment process of the parameters can refer to the technical details of the loss function such as cross entropy and the technical details of the optimization method such as gradient descent, which will not be repeated here. On this basis, the instantiated segmentation network is used to perform instantiated segmentation on the medical image to be tested, and the predicted pixel category of each pixel in the medical image to be tested can be obtained. For each bleeding focus category, the contour of the bleeding focus category can be obtained based on the connected domain formed by the pixels belonging to the bleeding focus category, thereby obtaining the contour and category of each bleeding focus.

[0047] In one implementation scenario, the detection result may further include an intermediate result in the process of calculating the amount of bleeding, such as at least one of the outline of the bleeding focus and the category of the bleeding focus.

[0048] Taking the medical image to be tested as an example of a brain scan, the bleeding focus may include at least one of the following: intraventricular hemorrhage (IVH), intraparenchymal hemorrhage (IPH), subarachnoid hemorrhage (SAH), epidural hemorrhage (EDH), and subdural hematoma (SDH). Other situations can be deduced by analogy, and examples are not given one by one here. In addition, the contour and category of the bleeding focus, and its specific acquisition process can refer to the aforementioned related description, which will not be repeated here.

[0049] In one implementation scenario, the detection result may further include the pixel value of the bleeding focus. Taking the medical image to be tested as a CT image as an example, the pixel value is the HU (Hounsfield Unit) value. It should be noted that the HU value reflects the degree of absorption of X-rays by the tissue. For details, please refer to the relevant technical details of the CT image, which will not be repeated here. On this basis, the average HU value of each bleeding focus can be counted to reflect the degree of absorption of X-rays by each bleeding focus.

[0050] It should be noted that, as mentioned above, the detection result can be obtained by detection using a detection model, and the detection model is obtained based on sample medical image training, and the sample medical image is at least annotated with a sample contour of the hemorrhage. Specifically, as mentioned above, the detection model may include a target detection network and a target segmentation network, that is, the target detection model and the target segmentation network may be trained separately, and after the training converges, the medical image to be tested is detected by the target detection network and the target segmentation network respectively to obtain the detection result, which can be specifically referred to in the above-mentioned related description, and will not be repeated here; or, the detection model may also include an instantiated segmentation network, that is, only the instantiated segmentation network may be trained, and after the training converges, the medical image to be tested is detected by the instantiated segmentation network to obtain the detection result, which can be specifically referred to in the above-mentioned related description, and will not be repeated here. It can be seen that in the image detection process, various key indicators that help assist doctors can be automatically detected, greatly reducing the doctor's operation time, and subsequently displaying the medical image to be tested and the detection results on the same screen, so that doctors can quickly understand the key indicators and organ images at a glance, especially in the case of emergency applications where every second counts, which helps assist doctors to make decisions quickly and accurately.

[0051] Step S12: Displaying the medical image to be tested and the test results on the same screen on the interactive interface.

[0052] In one implementation scenario, the interactive interface may include a first display area and a second display area, and as mentioned above, the medical image to be tested may be a three-dimensional medical image, then the first display area may be used to simultaneously display at least one two-dimensional medical image of the medical image to be tested on several preset planes, and the second display area may be used to display a list of bleeding foci, and the list of bleeding foci includes data items belonging to each bleeding foci, and the data items may include the amount of bleeding of the bleeding foci. The above method can improve the intuitiveness of the display of the test results by displaying the blood foci and their amount of bleeding detected in the medical image to be tested in the form of a list, and can display the medical image from different angles by displaying at least one two-dimensional medical image on several preset planes at the same time, which can improve the richness of the image display. Therefore, on the basis of improving the intuitiveness of the display of the test results and the richness of the image display, it can be helpful to further assist doctors in improving the speed of reading films.

[0053] In a specific implementation scenario, as mentioned above, the medical image to be tested may include but is not limited to a CT image. In addition, several preset planes may include but are not limited to: a coronal plane, a transverse plane, a sagittal plane, etc., which are not limited here. Specifically, each preset plane can display at least one two-dimensional medical image in the first display area. In addition, for the specific meanings of the above three planes, please refer to the relevant technical details, which will not be repeated here.

[0054] In a specific implementation scenario, the data items may also include the category and pixel value of the bleeding focus, which are not limited here. Taking the medical image to be tested as an example of a brain scan, the categories included in the data items may be at least one of the following: intraventricular hemorrhage (IVH), intracerebral parenchymal hemorrhage (IPH), subarachnoid hemorrhage (SAH), epidural hemorrhage (EDH), and subdural hematoma (SDH). It should be noted that in order to facilitate the display in the bleeding focus list, the category of the bleeding focus can be displayed with English abbreviations, and other situations can be deduced by analogy, and no examples are given here one by one. In addition, taking the medical image to be tested as a CT image as an example, the pixel value of the bleeding focus can be the HU value, which can reflect the degree of absorption of the bleeding focus to X-rays. The above method can display its category and amount of bleeding at the same time on the interactive interface, so it can improve the information richness of each bleeding focus, which is conducive to further assisting doctors to improve the reading speed.

[0055] In a specific implementation scenario, please refer to Figure 2 , Figure 2 Schematic diagram of an embodiment of an interactive interface. Figure 2 As shown, the dotted rectangular box on the right is the first display area, and the dotted rectangular box on the left is the second display area. The first display area on the right side of the interactive interface can display a two-dimensional medical image of the medical image to be tested on the three preset planes of the coronal plane, the transverse plane, and the sagittal plane. Specifically, the two-dimensional medical image of the medical image to be tested located in the middle layer of the coronal plane can be displayed by default in the first display area, and the two-dimensional medical image of the medical image to be tested located in the middle layer of the transverse plane can be displayed by default, and the two-dimensional medical image of the medical image to be tested located in the middle layer of the sagittal plane can be displayed by default. It should be noted that the user can also switch the two-dimensional medical images of other layers on each preset plane respectively. Taking the cross section as an example, the first display area can display the two-dimensional medical image of the medical image to be tested located on the N / 2th layer on the cross section by default, and the user can also switch to the first layer, second layer, third layer, etc. of the cross section. Two-dimensional medical images, which are not limited here. Other preset planes can be deduced by analogy, and examples are not given one by one here.

[0056] In a specific implementation scenario, please refer to Figure 3 , Figure 3Schematic diagram of an embodiment of the second display area. Figure 3 As shown, each row of the bleeding focus list can display a data item (as shown by the dashed rectangular box), and the data item may include the category, pixel value, bleeding amount, etc. of the bleeding focus. Specifically, Figure 3 Taking the medical image to be tested obtained by scanning the brain as an example, a possible situation of data item display is described exemplarily, where IPH indicates that the category of the bleeding focus is intraparenchymal hemorrhage, 66.7Hu indicates that the CT HU value of the bleeding focus is 66.7, and 20.5cc indicates that the amount of bleeding in the bleeding focus is 20.5 cubic centimeters. It should be noted that Figure 3 The above is only one implementation method for actual application scenarios, and does not limit the specific presentation method of data items.

[0057] In a specific implementation scenario, as mentioned above, the detection result also includes the contour of the bleeding focus. Based on this, in response to the data item selected by the user in the bleeding focus list, the contour of the bleeding focus to which the selected data item belongs can be used as the target contour, and the target contour can be marked in a preset style on the two-dimensional medical image displayed in the first display area. Figure 3 As shown, a selection box can be set for each data item in the bleeding focus list for the user to select, that is, when the user checks the selection box corresponding to a data item, it means that the data item has been selected. In addition, the preset styles may include but are not limited to: line color (such as red, green, etc.), line type (such as solid line, dotted line, etc.), line thickness (such as 20 pounds, 30 pounds, etc.), fill style (such as solid color fill, dot fill, etc.), fill transparency (such as 50% transparency, 60% transparency, etc.), which are not limited here. Please refer to Figure 2 and Figure 3 ,like Figure 3 In response to the user selecting the first data item in the bleeding focus list, the contour of the bleeding focus (i.e., IPH) to which the data item belongs can be used as the target contour, and then Figure 2 In the two-dimensional medical image, a preset style (such as Figure 2 The target contour is marked with a black solid line (indicated by the arrow in the middle). Other cases can be deduced by analogy, and no examples are given here one by one. In addition, it should be noted that the data items selected by the user in the bleeding focus list can be one, two, three, etc., that is, this embodiment does not limit the number of data items selected by the user in the bleeding focus list. The above method can respond to the doctor's operation to mark the contour that the doctor expects to see on the medical image, which is conducive to further assisting the doctor to improve the speed of reading the film.

[0058] In an implementation scenario, after the test results are obtained, a structured test report can also be formed based on the medical image to be tested and the test results. For example, the first page of the test report can list the various bleeding foci detected, as well as the category, bleeding volume, pixel value (such as HU value), etc. of each bleeding foci, and the subsequent pages of the test report can display two-dimensional medical images on several preset planes, and mark the contour area of ​​each bleeding foci on the two-dimensional medical image.

[0059] The above scheme detects the medical image to be tested and obtains the detection result of the bleeding focus in the medical image to be tested, and the detection result includes the amount of bleeding of the bleeding focus. On this basis, the medical image to be tested and the detection result are displayed on the same screen on the interactive interface. Therefore, it can save the doctor from manually estimating the amount of bleeding and other related indicators by reading the medical image to be tested, thereby greatly reducing the doctor's manual operation time during the film reading process, and then assisting the doctor to improve the film reading speed.

[0060] See also Figure 4 , Figure 4 This is a flow chart of another embodiment of the image detection method of the present application. In the disclosed embodiment, the detection result includes not only the amount of bleeding, but also the outline of the bleeding focus. For details, please refer to the relevant description in the aforementioned disclosed embodiment, which will not be repeated here. Specifically, the following steps may be included:

[0061] Step S41: Detect the medical image to be detected to obtain the detection result of the bleeding focus in the medical image to be detected.

[0062] For details, please refer to the relevant description in the aforementioned disclosed embodiments, which will not be repeated here.

[0063] Step S42: Displaying the medical image to be tested and the test results on the same screen on the interactive interface.

[0064] For details, please refer to the relevant description in the aforementioned disclosed embodiments, which will not be repeated here.

[0065] Step S43: In response to the user's operation of modifying the contour in the interactive interface, the bleeding amount of the bleeding focus is re-analyzed based on the modified contour, and the bleeding amount displayed in the interactive interface is updated to the bleeding amount obtained by the re-analysis.

[0066] In an implementation scenario, the outline of the bleeding focus is displayed in the interactive interface. Figure 2As well as the related descriptions in the aforementioned disclosed embodiments, they will not be repeated here. It should be noted that the two-dimensional medical images of the medical image to be tested on several preset planes can be displayed on the interactive interface, that is, the medical image to be tested is composed of several two-dimensional medical images on each preset plane. For example, for a medical image to be tested with a size of 256*256*32 (i.e., a length of 256, a width of 256, and a height of 32), it can be considered that it is composed of 32 two-dimensional medical images stacked in the height direction in the cross section, 256 two-dimensional medical images stacked in the width direction in the coronal plane, and 256 two-dimensional medical images stacked in the length direction in the sagittal plane. Other situations can be deduced by analogy, and examples are not given one by one here. On this basis, for each preset plane, the user can switch the two-dimensional medical image displayed on the interactive interface to support the user to modify the contour area of ​​the hemorrhage in the two-dimensional medical images on different preset planes.

[0067] In a specific implementation scenario, the interactive interface may be provided with an image switching control corresponding to each preset plane, so as to support the user to switch the two-dimensional medical images displayed on different preset planes through the image switching controls corresponding to different preset planes. For example, the image switching control may be set to a progress bar, and the display of the two-dimensional medical image may be switched by moving the slider on the progress bar; or, the image switching control may be set to a "+" or "-" button, and the display of the next two-dimensional medical image may be switched by clicking the "+" button, and the display of the previous two-dimensional medical image may be switched by clicking the "-" button. It should be noted that the above-mentioned types of controls are only the situations that may be used in actual applications, and the actual types of image switching controls are not limited thereto.

[0068] In a specific implementation scenario, as described in the aforementioned disclosed embodiment, the interactive interface may include a first display area and a second display area, and the first display area is used to simultaneously display at least one two-dimensional medical image of the medical image to be tested on several preset planes, and the second display area is used to display a bleeding focus list, the bleeding focus list includes data items belonging to each bleeding focus, the data items include but are not limited to the amount of bleeding of the bleeding focus, the category of the bleeding focus, the pixel value of the bleeding focus, etc., which are not limited here. Based on this, in response to the data item selected by the user in the bleeding focus list, the outline of the bleeding focus to which the selected data item belongs can be used as the target outline, and the target outline can be marked in a preset style on the two-dimensional medical image. In other words, the user can select the data item corresponding to the bleeding focus whose outline is desired to be modified in the bleeding focus category, and the outline of the bleeding focus can be marked in a preset style on the two-dimensional medical image displayed in the second display area, so that the user can modify the outline of the bleeding focus.

[0069] In one implementation scenario, the modified contour in the two-dimensional medical image on the preset plane can be obtained, and in response to the user's switching instruction, the next two-dimensional medical image is displayed on the interactive interface, and the above steps of obtaining the modified contour in the two-dimensional medical image on the preset plane and subsequent steps are re-executed until the modified contours in each two-dimensional medical image on the preset plane are obtained. On this basis, the amount of bleeding can be analyzed based on the modified contours in each two-dimensional medical image on the preset plane. The above method can support doctors to modify the contours on each two-dimensional medical image on the preset plane through image switching, which can reduce the complexity of contour modification operations and is conducive to further assisting doctors to improve the speed of reading films.

[0070] In a specific implementation scenario, the user can modify the contour area by erasing the contour, redrawing the contour, etc. on the two-dimensional medical image. For example, editing tools such as "erase" and "brush" can be provided in the interactive interface, and the user can erase the contour by using the "erase" tool and redraw the contour by using the "brush" tool.

[0071] In a specific implementation scenario, after completing the contour modification of a two-dimensional medical image, the user can switch to the next two-dimensional medical image through the image switching control to modify the contour of the next two-dimensional medical image. The specific implementation method of the image switching control can refer to the above-mentioned related description, which will not be repeated here.

[0072] In a specific implementation scenario, after the contour modification of each two-dimensional medical image on the preset plane is completed, the modified contours in each two-dimensional medical image on the preset plane can be stacked and combined into a three-dimensional contour of the bleeding focus, and based on the three-dimensional contour and the physical distance between adjacent pixels, the volume of the bleeding focus is calculated as the bleeding amount of the bleeding focus. Of course, the volume can also be multiplied by the blood density to obtain the mass of the bleeding focus as the bleeding amount of the bleeding focus.

[0073] In one implementation scenario, after the bleeding volume of the bleeding focus is re-analyzed, the bleeding volume of the bleeding focus can be updated to the re-analyzed bleeding volume on the interactive interface. Figure 3 For example, after the user selects an IPH bleeding focus in the bleeding focus list to modify the contour, the bleeding amount of the IPH bleeding focus can be re-analyzed, and the bleeding amount 20.5cc of the IPH bleeding focus in the bleeding focus list can be updated to the bleeding amount obtained by the re-analysis.

[0074] The above scheme can respond to the doctor's operation to modify the automatically detected contour on the medical image, and automatically re-analyze the amount of bleeding based on this, which is conducive to improving the doctor's autonomy in the process of reading the film, and can save the doctor from manually performing complex analysis and calculations. Therefore, it can improve autonomy while assisting doctors to improve the reading speed.

[0075] See also Figure 5 , Figure 5 1 is a schematic diagram of a framework of an embodiment of an image detection device 50 of the present application. The image detection device 50 comprises: a detection module 51 and a display module 52. The detection module 51 is used to detect the medical image to be detected and obtain the detection result of the bleeding focus in the medical image to be detected; wherein the detection result includes the amount of bleeding of the bleeding focus; and the display module 52 is used to display the medical image to be detected and the detection result on the same screen in the interactive interface.

[0076] The above scheme detects the medical image to be tested and obtains the detection result of the bleeding focus in the medical image to be tested, and the detection result includes the amount of bleeding of the bleeding focus. On this basis, the medical image to be tested and the detection result are displayed on the same screen on the interactive interface. Therefore, it can save the doctor from manually estimating the amount of bleeding and other related indicators by reading the medical image to be tested, thereby greatly reducing the doctor's manual operation time during the film reading process, and then assisting the doctor to improve the film reading speed.

[0077] In some disclosed embodiments, the interactive interface includes a first display area and a second display area, and the medical image to be tested is a three-dimensional medical image, wherein the first display area is used to simultaneously display at least one two-dimensional medical image of the medical image to be tested on several preset planes, and the second display area is used to display a bleeding foci list, wherein the bleeding foci list includes data items respectively belonging to each bleeding foci, and the data items include the amount of bleeding of the bleeding foci.

[0078] Therefore, by displaying the blood foci and their bleeding amounts detected in the medical image to be tested in the form of a list, the intuitiveness of the display of the test results can be improved, and by simultaneously displaying at least one two-dimensional medical image on several preset planes, the medical images can be displayed from different angles, which can improve the richness of the image display. Therefore, on the basis of improving the intuitiveness of the display of the test results and the richness of the image display, it can help further assist doctors to improve the speed of reading films.

[0079] In some disclosed embodiments, the detection result also includes the category of the bleeding focus, and the data item also includes the category of the bleeding focus.

[0080] Therefore, the detection results also include the category of bleeding foci, and on the basis of the data item including the amount of bleeding of the bleeding foci, it further includes the category of the bleeding foci, that is, for each bleeding foci, its category and amount of bleeding can be displayed simultaneously on the interactive interface, so the information richness of each bleeding foci can be improved, which is conducive to further assisting doctors to improve the speed of reading films.

[0081] In some disclosed embodiments, the detection result also includes the outline of the bleeding lesion, and the image detection device 50 also includes a selection module for responding to a data item selected by a user in the bleeding lesion list, and taking the outline of the bleeding lesion to which the selected data item belongs as the target outline; the image detection device 50 also includes a marking module for marking the target outline in a preset style on the two-dimensional medical image.

[0082] Therefore, the detection result also includes the outline of the bleeding focus, and in response to the data item selected by the user in the bleeding focus list, the outline of the bleeding focus to which the selected data item belongs is used as the target outline, and the target outline is marked in a preset style on the two-dimensional medical image. That is, it can respond to the doctor's operation to mark the outline that the doctor expects to see on the medical image, which is conducive to further assisting the doctor to improve the speed of reading the film.

[0083] In some disclosed embodiments, the detection result also includes the outline of the bleeding focus, and the image detection device 50 also includes an analysis module for responding to the user's operation of modifying the outline in the interactive interface, re-analyzing the amount of bleeding of the bleeding focus based on the modified outline, and updating the amount of bleeding displayed on the interactive interface to the amount of bleeding obtained by the re-analysis.

[0084] Therefore, being able to modify the automatically detected contours on the medical image in response to the doctor's operation and automatically re-analyze the amount of bleeding based on this can help improve the doctor's autonomy in the film reading process and save the doctor from having to perform complex manual analysis and calculations. Therefore, it can help doctors improve their reading speed while improving their autonomy.

[0085] In some disclosed embodiments, the medical image to be tested is a three-dimensional medical image, the medical image to be tested includes several two-dimensional medical images on a preset plane, and the interactive interface displays a two-dimensional medical image on the preset plane, the analysis module includes a modification submodule, which is used to obtain the modified contour in the two-dimensional medical image on the preset plane; the analysis module includes a switching submodule, which is used to respond to the user's switching instruction, display the next two-dimensional medical image on the interactive interface, and re-execute the step of obtaining the modified contour in the two-dimensional medical image on the preset plane and subsequent steps until the modified contours in each two-dimensional medical image on the preset plane are obtained; the analysis module includes an analysis submodule, which is used to analyze the amount of bleeding based on the modified contour area in each two-dimensional medical image on the preset plane.

[0086] Therefore, it can support doctors to modify the contours on various two-dimensional medical images of preset planes through image switching, which can reduce the complexity of contour modification operations and help further assist doctors to improve the speed of reading films.

[0087] In some disclosed embodiments, the detection result is obtained by detection using a detection model, and the detection model is trained based on sample medical images, and the sample medical images are at least annotated with sample contours of bleeding foci.

[0088] Therefore, the detection result is obtained by using the detection model, and the detection model is trained based on sample medical images, and the sample medical images are at least marked with sample contours of bleeding foci, so it can be beneficial to improve the efficiency of image detection.

[0089] In some disclosed embodiments, the medical image to be tested is obtained by scanning the brain, and the bleeding lesions include at least one of the following: intraventricular hemorrhage, intracerebral hemorrhage, subarachnoid hemorrhage, epidural hemorrhage, and subdural hematoma.

[0090] Therefore, the medical image to be tested is obtained by scanning the brain, and the bleeding foci include at least one of the following: intraventricular hemorrhage, intracerebral parenchymal hemorrhage, subarachnoid hemorrhage, epidural hemorrhage, and subdural hematoma, that is, it can support image detection of different types of cerebral hemorrhage.

[0091] In some disclosed embodiments, the detection result also includes the category of the bleeding focus; and / or, the detection result also includes the outline of the bleeding focus; and / or, the detection result also includes the pixel value of the bleeding focus.

[0092] Therefore, the detection result also includes at least one of the category and outline of the bleeding focus, and the pixel value of the bleeding focus, so it can be helpful to further enrich the auxiliary information that can be provided to doctors, and help further assist doctors to improve the speed of reading films.

[0093] See also Figure 6 , Figure 6 : is a schematic diagram of a framework of an embodiment of an electronic device 60 of the present application. The electronic device 60 includes a memory 61 and a processor 62 coupled to each other, and the processor 62 is used to execute program instructions stored in the memory 61 to implement the steps of any of the above-mentioned image detection method embodiments. In a specific implementation scenario, the electronic device 60 may include but is not limited to: a microcomputer, a server, and in addition, the electronic device 60 may also include a mobile device such as a laptop computer and a tablet computer, which is not limited here.

[0094] Specifically, the processor 62 is used to control itself and the memory 61 to implement the steps of any of the above-mentioned image detection method embodiments. The processor 62 can also be referred to as a CPU (Central Processing Unit). The processor 62 may be an integrated circuit chip with signal processing capabilities. The processor 62 may also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. In addition, the processor 62 may be implemented by an integrated circuit chip.

[0095] The above scheme detects the medical image to be tested and obtains the detection result of the bleeding focus in the medical image to be tested, and the detection result includes the amount of bleeding of the bleeding focus. On this basis, the medical image to be tested and the detection result are displayed on the same screen on the interactive interface. Therefore, it can save the doctor from manually estimating the amount of bleeding and other related indicators by reading the medical image to be tested, thereby greatly reducing the doctor's manual operation time during the film reading process, and then assisting the doctor to improve the film reading speed.

[0096] See also Figure 7 , Figure 7 The schematic diagram of the framework of an embodiment of a computer-readable storage medium 70 of the present application. The computer-readable storage medium 70 stores program instructions 701 that can be executed by a processor, and the program instructions 701 are used to implement the steps of any of the above-mentioned image detection method embodiments.

[0097] The above scheme detects the medical image to be tested and obtains the detection result of the bleeding focus in the medical image to be tested, and the detection result includes the amount of bleeding of the bleeding focus. On this basis, the medical image to be tested and the detection result are displayed on the same screen on the interactive interface. Therefore, it can save the doctor from manually estimating the amount of bleeding and other related indicators by reading the medical image to be tested, thereby greatly reducing the doctor's manual operation time during the film reading process, and then assisting the doctor to improve the film reading speed.

[0098] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0099] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0100] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

Claims

1. An image detection method, characterized in that: include: Detecting the medical image to be detected to obtain a detection result of a bleeding focus in the medical image to be detected; wherein the detection result includes the amount of bleeding and the outline of the bleeding focus; The medical image to be tested and the detection result are displayed on the same screen on an interactive interface; wherein the interactive interface includes a first display area and a second display area, and the medical image to be tested is a three-dimensional medical image, the first display area is used to simultaneously display at least one two-dimensional medical image of the medical image to be tested on a plurality of preset planes, and the second display area is used to display a bleeding focus list, the bleeding focus list includes data items respectively belonging to each of the bleeding focuses, and the data items include the amount of bleeding of the bleeding focus; In response to the user's operation of modifying the contour on the interactive interface, the modified contour in the two-dimensional medical image on the preset plane is obtained, and in response to the user's switching instruction, the next two-dimensional medical image is displayed on the interactive interface, and the step of obtaining the modified contour in the two-dimensional medical image on the preset plane and subsequent steps are re-executed until the modified contours in each of the two-dimensional medical images on the preset plane are obtained, and based on the modified contours in each of the two-dimensional medical images on the preset plane, the bleeding amount is analyzed and obtained, and the bleeding amount displayed on the interactive interface is updated to the bleeding amount obtained by the re-analysis.

2. The method according to claim 1, characterized in that The detection result further includes the category of the bleeding focus, and the data item further includes the category of the bleeding focus.

3. The method according to claim 1, characterized in that The detection result also includes the outline of the bleeding focus, and the method further includes: In response to a data item selected by a user in the bleeding focus list, taking the contour of the bleeding focus to which the selected data item belongs as a target contour; The target contour is marked on each of the two-dimensional medical images in a preset style.

4. The method according to any one of claims 1 to 3, characterized in that: The detection result is obtained by detection using a detection model, and the detection model is trained based on sample medical images, and the sample medical images are at least marked with sample contours of bleeding foci.

5. The method according to any one of claims 1 to 3, characterized in that: The medical image to be tested is obtained by scanning the brain, and the bleeding focus includes at least one of the following: intraventricular hemorrhage, cerebral parenchymal hemorrhage, subarachnoid hemorrhage, epidural hemorrhage, and subdural hematoma.

6. The method according to any one of claims 1 to 3, characterized in that: The test result also includes the category of the bleeding focus; And / or, the detection result also includes the pixel value of the bleeding focus.

7. An image detection device, characterized in that: include: A detection module, used to detect the medical image to be detected, and obtain a detection result of the bleeding focus in the medical image to be detected; wherein the detection result includes the bleeding amount and contour of the bleeding focus; A display module, used for displaying the medical image to be tested and the detection result on the same screen on an interactive interface; wherein the interactive interface includes a first display area and a second display area, and the medical image to be tested is a three-dimensional medical image, the first display area is used for simultaneously displaying at least one two-dimensional medical image of the medical image to be tested on a plurality of preset planes, and the second display area is used for displaying a bleeding focus list, the bleeding focus list includes data items respectively belonging to each of the bleeding focuses, and the data items include the amount of bleeding of the bleeding focus; An analysis module is used to respond to a user's operation of modifying the contour on the interactive interface, obtain the modified contour in the two-dimensional medical image on the preset plane, respond to a user's switching instruction, display the next two-dimensional medical image on the interactive interface, re-execute the step of obtaining the modified contour in the two-dimensional medical image on the preset plane and subsequent steps until the modified contours in each of the two-dimensional medical images on the preset plane are obtained, analyze the bleeding amount based on the modified contours in each of the two-dimensional medical images on the preset plane, and update the bleeding amount displayed on the interactive interface to the bleeding amount obtained by the re-analysis.

8. An electronic device, characterized in that: It comprises a memory and a processor coupled to each other, wherein the processor is used to execute program instructions stored in the memory to implement the image detection method according to any one of claims 1 to 6.

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

Images