A method and device for determining a surgical path, electronic equipment and storage medium

Abstract

The application discloses a surgical path determination method and device, electronic equipment and storage medium. A lesion image and an angiogram image of a lesion site are acquired; a plurality of candidate surgical starting points are determined on the edge contour line of the lesion image, and a plurality of candidate surgical paths are determined based on the plurality of candidate surgical starting points and a corresponding surgical endpoint of the lesion site; first path feature information corresponding to each candidate surgical path is determined based on the lesion image, and second path feature information corresponding to each candidate surgical path is determined based on the angiogram image; and a target neural network model obtained through pre-training, the first path feature information and the second path feature information corresponding to the candidate surgical path are used to determine a target surgical path corresponding to the lesion site from the plurality of candidate surgical paths, thereby solving the problems of low efficiency and low accuracy of surgical path determination and improving the efficiency and accuracy of surgical path determination.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a surgical path determination method and device, electronic equipment and a storage medium. BACKGROUND

[0002] Under the premise that the lesion site in the body is known, the line segment between the surgical starting point on the body surface and the surgical ending point of the lesion site is the surgical path. If the surgical path passes through a dangerous area that cannot be touched by the human body, it will cause greater harm to the patient, and is not conducive to the operation, so it is necessary to determine the surgical path in advance.

[0003] At present, the usual way to determine the surgical path is for the reception personnel to comprehensively analyze the image data and clinical data of the patient according to their own experience, and then design the surgical path in the surgical planning software. However, this approach has the problem of high dependence on manual work, which requires the reception personnel to determine the surgical path based on their past experience, which is subjective and has the technical problems of low efficiency and low accuracy in determining the surgical path. SUMMARY

[0004] The embodiments of the present application provide a surgical path determination method and device, electronic equipment and a storage medium to improve the efficiency and accuracy of determining the surgical path.

[0005] In a first aspect, the present application provides a surgical path determination method, which comprises:

[0006] obtaining a lesion image containing the same lesion site and an angiogram corresponding to the lesion image;

[0007] determining a plurality of candidate surgical starting points on the edge contour line of the lesion image, and determining a plurality of candidate surgical paths based on the plurality of candidate surgical starting points and the surgical ending point corresponding to the lesion site;

[0008] determining first path feature information corresponding to each candidate surgical path based on the lesion image, and determining second path feature information corresponding to each candidate surgical path based on the angiogram;

[0009] determining a target surgical path corresponding to the lesion site from the plurality of candidate surgical paths based on a target neural network model trained in advance, the first path feature information and the second path feature information corresponding to the candidate surgical path.

[0010] In a second aspect, the present application provides a surgical path determination device, which comprises:

[0011] an image data acquisition module for acquiring a lesion image containing the same lesion site and an angiogram corresponding to the lesion image;

[0012] The candidate path determination module is configured to determine a plurality of candidate surgical starting points on the edge contour line of the lesion image, and determine a plurality of candidate surgical paths based on the plurality of candidate surgical starting points and a surgical ending point corresponding to the lesion site;

[0013] The feature information determination module is configured to determine first path feature information corresponding to each candidate surgical path based on the lesion image, and determine second path feature information corresponding to each candidate surgical path based on the angiographic image;

[0014] The target path determination module is configured to determine a target surgical path corresponding to the lesion site from the plurality of candidate surgical paths based on a pre-trained target neural network model, the first path feature information and the second path feature information of the candidate surgical paths.

[0015] In a third aspect, the present application provides an electronic device, comprising:

[0016] at least one processor; and

[0017] a memory connected with the at least one processor in communication; wherein,

[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the surgical path determination method of any one of the embodiments of the present application.

[0019] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions, and the computer instructions are used to enable a processor to implement the surgical path determination method of any one of the embodiments of the present application when executed.

[0020] The technical scheme provided by the embodiment of the present application comprises the following steps: obtaining a lesion image and an angiogram of a lesion site, then determining a plurality of candidate surgical starting points on the edge contour line of the lesion image, and determining a plurality of candidate surgical paths based on the plurality of candidate surgical starting points and a surgical endpoint corresponding to the lesion site. The first path feature information corresponding to each candidate surgical path is determined based on the lesion image, and the second path feature information corresponding to each candidate surgical path is determined based on the angiogram. Then, the target surgical path corresponding to the lesion site is determined from the plurality of candidate surgical paths based on the target neural network model, the first path feature information and the second path feature information of the candidate surgical path, so as to realize automatic determination of the surgical path without human intervention. The first path feature information and the second path feature information are obtained by parameterizing the candidate surgical path in the lesion image and the angiogram, and the suitable surgical path is accurately determined by the target neural network model for path prediction of the first path feature information and the second path feature information, thereby improving the determination efficiency and accuracy of the surgical path.

[0021] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0023] Figure 1 The flowchart of a surgical path determination method provided for the first embodiment of the present application;

[0024] Figure 2 The lesion image schematic diagram related to the first embodiment of the present application;

[0025] Figure 3 The schematic diagram of the candidate surgical path in the angiogram related to the first embodiment of the present application;

[0026] Figure 4 The flowchart of a surgical path determination method provided for the second embodiment of the present application;

[0027] Figure 5 The functional tissue area schematic diagram related to the second embodiment of the present application;

[0028] Figure 6A structural schematic diagram of a surgical path determination device provided for the third embodiment of the present application is shown in the figure.

[0029] Figure 7 A structural schematic diagram of an electronic device provided for the fourth embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0030] In order to make the personnel in the art better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the figures in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person of ordinary skill in the art without creative labor should belong to the scope of protection of the present application.

[0031] It should be noted that the terms "first preset condition", "second preset condition" and the like in the specification and claims of the present application and the above-mentioned figures are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or electronic device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or electronic devices.

[0032] Embodiment one

[0033] Figure 1 A flowchart of a surgical path determination method provided for the first embodiment of the present application is shown in the figure. The present embodiment can be applicable to the case of determining a surgical path under the premise that the lesion site of the patient is known. The method can be executed by a surgical path determination device, which can be realized in the form of hardware and / or software. The device can be configured on a computer device, which can be a notebook, a desktop computer, a smart tablet and the like. As shown in the figure, the method comprises the following steps. Figure 1

[0034] S110, obtaining a lesion image and an angiogram image of a lesion site.

[0035] ​The lesion image is an examination image taken by the patient during the examination, and the lesion image contains a lesion target position. For example, the lesion image can be a magnetic resonance image (MRI). The angiography image is an examination image showing the structure of blood vessels. In this embodiment, the angiography image contains a lesion target point consistent with the lesion image. For example, the angiography image can be a CT angiography (CTA).

[0036] In actual production, the patient needs to undergo a series of preoperative examinations before surgery. The preoperative examination items include taking the lesion image and the angiography image. The lesion image and the angiography image can be calibrated and fused using existing medical software, so that the lesion image and the angiography image can truly reflect the actual physiological structure of the lesion site and the surrounding tissues and organs of the patient.

[0037] For example, for a patient with cerebral hemorrhage, a series of preoperative examinations are required before surgery to determine whether the patient can undergo surgery and to determine the lesion target position of the surgery. The results of these preoperative examinations include the lesion image and the angiography image. The lesion image and the angiography image are three-dimensional images. After determining the lesion target position, a two-dimensional image with the most obvious lesion target point can be further determined from the three-dimensional lesion image as the lesion image. And determine the two-dimensional angiography image corresponding to the two-dimensional lesion image.

[0038] S120, determining a plurality of candidate surgical starting points on the edge contour line of the lesion image, and determining a plurality of candidate surgical paths based on the plurality of candidate surgical starting points and the surgical endpoint corresponding to the lesion site.

[0039] The edge contour line is a line formed by edge contour pixel points in the lesion image. The edge contour line represents the body surface in a physiological sense. When performing surgery, the lesion target point is the surgical endpoint, and the problem is how to determine the surgical starting point. Therefore, the candidate surgical starting point is a series of hypothetical surgical starting points. In this embodiment, the candidate surgical path is a straight line segment determined by the surgical endpoint corresponding to the lesion site and the candidate surgical starting point.

[0040] Specifically, based on the obtained lesion image, the edge detection model can detect and mark the edge contour line of the lesion image, so as to determine a plurality of candidate surgical starting points on the edge contour line. Each determined candidate surgical starting point is connected to the surgical endpoint corresponding to the lesion site by a straight line, thereby determining a candidate surgical path corresponding to each candidate surgical starting point.

[0041] For example, the lesion image schematic diagram is shown in Figure 2 . As Figure 2As shown, the outer curve is the edge contour line, and four candidate surgical starting points A1, A2, A3 and A4 are determined on the edge contour line, and O is the corresponding surgical endpoint of the lesion site. L1 is a candidate surgical path determined by the candidate surgical starting point A1 and the corresponding surgical endpoint O of the lesion site. Correspondingly, L2 is a candidate surgical path determined by the candidate surgical starting point A2 and the surgical endpoint O; L3 is a candidate surgical path determined by the candidate surgical starting point A3 and the surgical endpoint O; and L4 is a candidate surgical path determined by the candidate surgical starting point A4 and the surgical endpoint O.

[0042] In S130, first path feature information corresponding to each candidate surgical path is determined based on the lesion image, and second path feature information corresponding to each candidate surgical path is determined based on the angiogram image.

[0043] The first path feature information is used to represent the data features of the candidate surgical path on the lesion image, and the second path feature information is used to represent the data features of the candidate surgical path on the angiogram image. Optionally, the first path feature information and the second path feature information can be in the form of a feature vector. Each candidate surgical path determines corresponding first path feature information and second path feature information.

[0044] Specifically, each candidate surgical path in the lesion image can be mapped to the angiogram image. For each candidate surgical path, based on the pixel points through which the candidate surgical path passes in the lesion image, the first path feature parameter corresponding to the candidate surgical path is determined; and based on the pixel points through which the candidate surgical path passes in the angiogram image, the second path feature parameter corresponding to the candidate surgical path is determined.

[0045] In S140, based on the target neural network model pre-trained, the first path feature information and the second path feature information corresponding to the candidate surgical path, the target surgical path corresponding to the lesion site is determined from the plurality of candidate surgical paths.

[0046] The target neural network model is pre-trained. The training sample set used in the training process of the target neural network model includes at least one sample data, and each sample data includes a lesion image of the same lesion site, an angiogram image corresponding to the lesion image, and a pre-labeled surgical path. Since the training sample set is historical diagnosis and treatment data, the pre-labeled surgical path is determined according to the real surgical path determined by the reception personnel.

[0047] The target surgical path is the final surgical path determined from the plurality of candidate surgical paths. The target surgical path is used to represent the relatively ideal surgical path in the candidate surgical paths.

[0048] Specifically, the first path feature information and the second path feature information corresponding to each of the candidate surgical paths are taken as a group of feature vectors, and input to a target neural network model trained in advance. The target neural network model can output a prediction probability value corresponding to each of the candidate surgical paths. Then, according to the prediction probability value corresponding to each of the candidate surgical paths, a target surgical path corresponding to the lesion site is determined from the plurality of candidate surgical paths.

[0049] On the basis of the above examples, the first path feature information M1 and the second path feature information N1 of L1 can be taken as a group of feature vectors (M1, N1). Correspondingly, the feature vector corresponding to L2 can be represented as (M2, N2); the feature vector corresponding to L3 can be represented as (M3, N3); and the feature vector corresponding to L4 can be represented as (M4, N4). Then, (M1, N1), (M2, N2), (M2, N2) and (M2, N2) are input to the target neural network model, and the target neural network model outputs a prediction probability value X1 corresponding to L1, a prediction probability value X2 corresponding to L2, a prediction probability value X3 corresponding to L3, and a prediction probability value X4 corresponding to L4. Finally, according to the numerical size relationship of X1, X2, X3 and X4, a target surgical path corresponding to the lesion site is determined from L1, L2, L3 and L4.

[0050] It should be particularly noted that a plurality of target neural network models corresponding to different lesion sites can be trained in advance. For example, a target neural network model corresponding to a head lesion site, a target neural network model corresponding to an abdominal lesion site, etc. can be trained in advance. In the specific application process, according to the corresponding lesion site of the lesion site, a target neural network model corresponding to the lesion site is determined, and further, the first path feature information and the second path feature information corresponding to the lesion site are input to the target neural network model corresponding to the lesion site, so as to obtain a target surgical path corresponding to the lesion site.

[0051] The technical scheme provided by the embodiment of the present application comprises the following steps: obtaining a lesion image and an angiogram of a lesion site, then determining a plurality of candidate surgical starting points on the edge contour line of the lesion image, and determining a plurality of candidate surgical paths based on the plurality of candidate surgical starting points and the surgical endpoint corresponding to the lesion site. The first path feature information corresponding to each candidate surgical path is determined based on the lesion image, and the second path feature information corresponding to each candidate surgical path is determined based on the angiogram. Then, the target surgical path corresponding to the lesion site is determined from the plurality of candidate surgical paths based on the target neural network model, the first path feature information and the second path feature information of the candidate surgical path, so as to realize automatic determination of the surgical path without human intervention. In addition, the first path feature information and the second path feature information are obtained by parameterizing the candidate surgical path in the lesion image and the angiogram, and the suitable surgical path can be accurately determined by performing path prediction on the first path feature information and the second path feature information through the target neural network model, thereby improving the determination efficiency and accuracy of the surgical path.

[0052] On the basis of the above-mentioned embodiments, the plurality of candidate surgical starting points on the edge contour line of the lesion image are determined, specifically comprising: determining a candidate surgical starting point every preset number of pixel points on the edge contour line of the lesion image; or determining the spacing information between the adjacent two candidate surgical starting points based on the length of the edge contour line and the preset number, and determining the plurality of candidate surgical starting points on the edge contour line of the lesion image based on the spacing information.

[0053] In the present embodiment, the plurality of candidate surgical starting points on the edge contour line of the lesion image can include at least two ways. One way is to determine a candidate surgical starting point every preset number of pixel points on the edge contour line of the lesion image, for example, a candidate surgical starting point can be determined every 20 pixel points. Another way is to determine a specified number of candidate surgical starting points on the edge contour line. For example, it is predetermined to determine 100 candidate surgical starting points on the edge contour line, so the preset number is 100. The length of the edge contour line can be directly determined according to the pixel points corresponding to the edge contour line, and then the length of the edge contour line divided by the preset number can determine the spacing information between the adjacent two candidate surgical starting points, so that the plurality of candidate surgical starting points on the edge contour line of the lesion image can be determined according to the spacing information.

[0054] Embodiment two

[0055] Figure 4 The flowchart of the surgical path determination method provided by the second embodiment of the present application is based on the above-mentioned embodiments, and the second embodiment of the present application is further refined on the basis of the above-mentioned embodiments S130, and can be combined with one or more optional schemes in the above-mentioned embodiments. For example, Figure 4As shown, the method comprises:

[0056] S210, acquiring a lesion image and an angiogram image of the lesion site.

[0057] S220, determining a plurality of candidate surgical starting points on the edge contour line of the lesion image, and determining a plurality of candidate surgical paths based on the plurality of candidate surgical starting points and a surgical endpoint corresponding to the lesion site.

[0058] S231, for each candidate surgical path, determining a target functional tissue area through which the candidate surgical path passes in the lesion image and a first number of pixel points corresponding to each target functional tissue area based on a plurality of functional tissue areas corresponding to the target site and the lesion image.

[0059] Wherein, the target site is a body part where the lesion site is located, for example, the target site can be the head, the abdomen. The functional tissue area is a pre-set different functional partition corresponding to the target site. The target functional tissue area is the functional tissue area through which the candidate surgical path passes in the lesion image.

[0060] Optionally, if the lesion site is a head lesion site, the functional tissue area corresponding to the head lesion site includes: an epidermal skull area and a brain function classification area; if the lesion site is an abdominal lesion site, the functional tissue area corresponding to the abdominal lesion site includes: a skin, connective tissue and organ classification area.

[0061] Wherein, the brain function partition includes at least one sub-function partition. The organ classification area includes at least one sub-organ partition. For example, see Figure 5 The lesion image corresponding to the head is shown as Figure 5 The target site is the head, and the functional tissue area corresponding to the head includes the skull area S1, the sub-function partition S2, the sub-function partition S3, the sub-function partition S4, the sub-function partition S5 and the sub-function partition S6.

[0062] Optionally, the weight value corresponding to the epidermal skull area is less than the weight value corresponding to the brain function classification area; the weight value corresponding to the skin is less than the weight value corresponding to the connective tissue, and the weight value corresponding to the connective tissue is less than the weight value corresponding to the organ classification area.

[0063] In this embodiment, different weight values are pre-set for different functional tissue areas. During the operation, if the surgical path passes through the epidermal skull area, it has little effect on the patient, but if the surgical path passes through the brain function classification area corresponding to the important brain tissue, it will affect the operation effect. Based on this, the weight value corresponding to the epidermal skull area is less than the weight value corresponding to the brain function classification area. The sub-function partitions included in the brain function classification area are set with different weight values according to the importance. For example, the more important the sub-function partition, the greater the weight value.

[0064] If the lesion site is an abdominal lesion site, the functional tissue area includes: skin, connective tissue and organ classification area. Different weight values can be set for the skin, connective tissue and organ classification area respectively. During the operation, the operation path passing through the skin will not affect the operation, passing through the connective tissue may affect the operation, and the operation path passing through important organ tissue will have a serious negative impact on the operation. Based on this, the weight value corresponding to the skin is less than the weight value corresponding to the connective tissue, and the weight value corresponding to the connective tissue is less than the weight value corresponding to the organ classification area. The sub-organ partition included in the organ classification area sets different weight values according to the importance.

[0065] It should be particularly pointed out that the method for determining the first path feature information of each candidate operation path is the same, and one of the candidate operation paths will be exemplarily described below.

[0066] In this embodiment, the plurality of functional tissue areas corresponding to the target site can be pre-marked. According to the target site where the lesion site is located, the plurality of functional tissue areas corresponding to the target site can be retrieved. Further, for a candidate operation path, first determine that the candidate operation path passes through at least one target functional tissue area in the lesion image. Further determine the first number of pixel points passed through by the candidate operation path in each target functional tissue area.

[0067] Exemplarily, as shown in Figure 5 The target functional tissue areas passed through by the candidate operation path L1 include S1, S2, S5 and S6, the first number of pixel points passed through by the target functional tissue area S1 is 100, the first number of pixel points passed through by the target functional tissue area S2 is 500, the first number of pixel points passed through by the target functional tissue area S5 is 150, and the first number of pixel points passed through by the target functional tissue area S6 is 130.

[0068] S232, based on the correspondence between the target functional tissue area and the weight value of the functional tissue area, determine the target weight value corresponding to each target functional tissue area.

[0069] Among them, the target weight value is the weight value corresponding to the target functional tissue area.

[0070] In actual application, the correspondence between different functional tissue areas and weight values can be pre-set, and on the basis of determining the target functional tissue area, the target weight value corresponding to each target functional tissue area is determined according to the correspondence.

[0071] On the basis of the above examples, the target weight value q1 corresponding to the target functional tissue area S1, the target weight value q2 corresponding to the target functional tissue area S2, the target weight value q3 corresponding to the target functional tissue area S5, and the target weight value q4 corresponding to the target functional tissue area S6.

[0072] S233, determining a functional area characteristic value corresponding to each target functional tissue area based on the first number of passing pixel points and the target weight value.

[0073] In this embodiment, the functional area characteristic value corresponding to each target functional tissue area is the product of the first number of passing pixel points and the target weight value.

[0074] On the basis of the above examples, the functional area characteristic value corresponding to the target functional tissue area S1 can be represented as 100 x q1, the functional area characteristic value corresponding to the target functional tissue area S2 can be represented as 500 x q2, the functional area characteristic value corresponding to the target functional tissue area S5 can be represented as 150 x q3, and the functional area characteristic value corresponding to the target functional tissue area S6 can be represented as 130 x q4.

[0075] S234, determining first path characteristic information corresponding to each candidate surgical path based on the functional area characteristic value corresponding to each target functional tissue area.

[0076] In this embodiment, the first path characteristic information corresponding to one candidate surgical path is a vector composed of the functional area characteristic values corresponding to the target functional tissue areas. On the basis of the above examples, the first path characteristic information corresponding to the candidate surgical path L1 can be represented as {100 x q1, 500 x q2, 150 x q3, 130 x q4}.

[0077] S241, determining each second passing pixel point of the candidate surgical path passing through the angiogram image, and determining a target blood vessel pixel point closest to each second passing pixel point.

[0078] The second passing pixel point is a pixel point corresponding to the candidate surgical path on the angiogram image. For example, as shown in FIG. 2B, the pixel points corresponding to the candidate surgical path L1 on the angiogram image are the second passing pixel points. Figure 3

[0079] In this embodiment, on the basis of the determination of the second passing pixel points, the target blood vessel pixel point closest to each second passing pixel point is further determined.

[0080] S242, for each second passing pixel point, determining blood vessel distance information corresponding to each second passing pixel point based on the second passing pixel point and the target blood vessel pixel point closest to the second passing pixel point.

[0081] ​The blood vessel distance information is actual physical length information between the second passing pixel point and the target blood vessel pixel point.

[0082] In this embodiment, the method for determining the blood vessel distance information corresponding to each second passing pixel point is the same. For one of the second passing pixel points, after determining the target blood vessel pixel point corresponding thereto, the blood vessel distance information corresponding to the second passing pixel point can be determined based on the distance between the second passing pixel point and the target blood vessel pixel point on the angiogram image and the imaging scale of the angiogram image.

[0083] S243, determining second path feature information corresponding to each candidate surgical path based on the blood vessel distance information.

[0084] In this embodiment, the second path feature information corresponding to one candidate surgical path is a vector composed of the blood vessel distance information.

[0085] Based on the above example, if the candidate surgical path L1 contains 10 second passing pixel points, and the blood vessel distance information corresponding to each second passing pixel point is d1, d2, d3, d4, d5, d6, d7, d8, d9 and d10 respectively, the second path feature information corresponding to the candidate surgical path L1 can be represented as {d1, d2, d3, d4, d5, d6, d7, d8, d9, d10}.

[0086] S250, inputting the first path feature information and the second path feature information corresponding to each candidate surgical path into a target neural network model trained in advance to perform path prediction, and obtaining a prediction probability value corresponding to each candidate surgical path.

[0087] In actual application, the first path feature information and the second path feature information corresponding to each candidate surgical path are taken as input quantities, and the input quantities are input into the target neural network model trained in advance. The output quantity of the target neural network model is a prediction probability value corresponding to each candidate surgical path.

[0088] S260, determining a target surgical path corresponding to a lesion site from a plurality of candidate surgical paths based on the prediction probability value.

[0089] In this embodiment, after determining the prediction probability value corresponding to each candidate surgical path, the candidate surgical path with the maximum prediction probability value is further determined as the target surgical path.

[0090] The technical scheme provided by the embodiment of the application, when determining the first path feature information corresponding to the to-be-selected surgical path, determines the target functional tissue area and the first number of pixel points corresponding to each target functional tissue area through which the to-be-selected surgical path passes through the lesion image based on the multiple functional tissue areas and the lesion image corresponding to the target area, determines the target weight value corresponding to each target functional tissue area based on the corresponding relationship between the target functional tissue area and the weight value, and determines the functional area feature value corresponding to each target functional tissue area based on the first number of pixel points and the target weight value, and then determines the first path feature information corresponding to each to-be-selected surgical path based on the functional area feature values corresponding to the target functional tissue areas. When determining the second path feature information corresponding to the to-be-selected surgical path, first, the second pixel points through which the to-be-selected surgical path passes through the angiogram image are determined, and the target blood vessel pixel point closest to each second pixel point is determined, and then for each second pixel point, the blood vessel distance information corresponding to each second pixel point is determined based on the second pixel point and the target blood vessel pixel point closest to the second pixel point, so as to determine the second path feature information corresponding to each to-be-selected surgical path based on the blood vessel distance information. In the process of determining the surgical path, the first path feature information of the to-be-selected surgical path in the lesion image and the second path feature information of the to-be-selected surgical path in the angiogram image are determined first, the parameterized evaluation of the image content is realized, the first path feature information and the second path feature information are predicted by the target neural network model, so as to determine the target surgical path, and the efficiency and accuracy of determining the surgical path are further improved.

[0091] Embodiment three

[0092] Figure 6 A structural schematic diagram of a surgical path determination device provided by the third embodiment of the application, which can execute the surgical path determination method provided by the embodiments of the application. The device comprises an image data acquisition module 310, a to-be-selected path determination module 320, a feature information determination module 330, and a target path determination module 340.

[0093] The image data acquisition module 310 is configured to acquire the lesion image and the angiogram image of the lesion site.

[0094] The to-be-selected path determination module 320 is configured to determine multiple to-be-selected surgical starting points on the edge contour line of the lesion image, and determine multiple to-be-selected surgical paths based on the multiple to-be-selected surgical starting points and the surgical endpoint corresponding to the lesion site.

[0095] The feature information determination module 330 is configured to determine the first path feature information corresponding to each to-be-selected surgical path based on the lesion image, and determine the second path feature information corresponding to each to-be-selected surgical path based on the angiogram image.

[0096] The target path determination module 340 is configured to determine the target surgical path corresponding to the lesion site from the plurality of candidate surgical paths based on the target neural network model pre-trained, the first path feature information and the second path feature information corresponding to the candidate surgical path.

[0097] On the basis of the above technical solutions, the candidate path determination module 320 comprises a candidate surgical starting point determination unit configured to determine a candidate surgical starting point on the edge contour line of the lesion image every preset number of pixel points; or, based on the length of the edge contour line and the preset number, determine spacing information between adjacent two candidate surgical starting points, and determine a plurality of candidate surgical starting points on the edge contour line of the lesion image based on the spacing information.

[0098] On the basis of the above technical solutions, the feature information determination module 330 comprises a first feature determination unit and a second feature determination unit. The first feature determination unit comprises:

[0099] The number of passes determination sub-unit is configured to, for each candidate surgical path, determine a first number of pixel points of a target functional tissue region through which the candidate surgical path passes the lesion image and each target functional tissue region based on the plurality of functional tissue regions corresponding to the target site and the lesion image;

[0100] The target weight value determination sub-unit is configured to determine a target weight value corresponding to each target functional tissue region based on the corresponding relationship between the target functional tissue region and the weight value;

[0101] The feature value determination sub-unit is configured to determine a functional region feature value corresponding to each target functional tissue region based on the first number of pixel points of passing and the target weight value;

[0102] The first feature determination sub-unit is configured to determine first path feature information corresponding to each candidate surgical path based on the functional region feature value corresponding to each target functional tissue region.

[0103] The second feature determination unit comprises:

[0104] The pixel point determination sub-unit is configured to determine each second passing pixel point of the candidate surgical path passing through the angiogram image, and determine a target blood vessel pixel point closest to each second passing pixel point;

[0105] The distance information determination sub-unit is configured to, for each second passing pixel point, determine blood vessel distance information corresponding to each second passing pixel point based on the second passing pixel point and the target blood vessel pixel point closest to the second passing pixel point;

[0106] The second feature determination subunit is configured to determine second path feature information corresponding to each candidate surgical path based on the blood vessel distance information.

[0107] On the basis of the above technical solutions, the target path determination module 340 comprises:

[0108] The prediction probability value determination subunit is configured to input the first path feature information and the second path feature information corresponding to each candidate surgical path into a target neural network model trained in advance to perform path prediction, and obtain a prediction probability value corresponding to each candidate surgical path.

[0109] The target path determination subunit is configured to determine a target surgical path corresponding to the lesion site from the plurality of candidate surgical paths based on the prediction probability value.

[0110] The technical solutions provided in the embodiments of the present application can achieve automatic determination of a surgical path without human intervention, and can improve the determination efficiency and accuracy of the surgical path.

[0111] The surgical path determination device provided in the embodiments of the present application can perform the surgical path determination method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0112] It should be noted that each unit and module included in the above device is only divided according to the function logic, but is not limited to the above division, as long as the corresponding function can be realized; in addition, the specific name of each functional unit is only for convenient mutual distinction, and is not used to limit the protection scope of the embodiments of the present application.

[0113] Embodiment Four

[0114] Figure 7A structural diagram of an electronic device is provided for Embodiment Four of the present application. The electronic device 10 is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart phones, wearable electronic devices (e.g., headsets, eyewear, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the present application as described and / or claimed in this document.

[0115] As shown in Figure 7 The electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., connected in communication with the at least one processor 11, where the memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 12 or loaded into the random access memory (RAM) 13 from the storage unit 18. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 13. An input / output (I / O) interface 15 is also connected to the bus 13.

[0116] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc., an output unit 17, such as various types of displays, a speaker, etc., a storage unit 18, such as a magnetic disk, an optical disk, etc., and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other electronic devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0117] The processor 11 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the surgical path determination method.

[0118] In some embodiments, the method of determining a surgical path can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 18. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 10 via, e.g., ROM 12 and / or communication unit 19. When the computer program is loaded onto RAM 13 and executed by processor 11, one or more steps of the method of determining a surgical path described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method of determining a surgical path by other means, e.g., with the aid of firmware.

[0119] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0120] Computer programs used to implement the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0121] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or electronic device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or electronic device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of electronic communication, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0122] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0123] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), blockchain networks, and the Internet.

[0124] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service. It should be understood that the various forms of flow shown above can be reordered, added, or deleted steps. For example, each step described in the present application can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein. The above specific embodiments do not constitute a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for determining a surgical path, characterized in that, include: Acquire images of the lesion and angiography corresponding to the lesion site; Multiple candidate surgical starting points are determined on the edge contour line of the lesion image, and multiple candidate surgical paths are determined based on the multiple candidate surgical starting points and the surgical ending points corresponding to the lesion sites. Based on the lesion image, determine the first path feature information corresponding to each candidate surgical path, and based on the angiography image, determine the second path feature information corresponding to each candidate surgical path. Based on the pre-trained target neural network model, the first path feature information and the second path feature information corresponding to the candidate surgical path, the target surgical path corresponding to the lesion site is determined from multiple candidate surgical paths; The edge contour line is the line formed by the edge contour pixels in the lesion image; The determination of the second path feature information corresponding to each candidate surgical path based on angiography images includes: Determine each second crossing pixel point of the angiography image through which the candidate surgical path passes, and determine the target blood vessel pixel point closest to each second crossing pixel point; For each second passing pixel, based on the second passing pixel and the target blood vessel pixel closest to the second passing pixel, determine the blood vessel distance information corresponding to each second passing pixel; Based on the blood vessel distance information, the second path feature information corresponding to each candidate surgical path is determined; The step of determining the first path feature information corresponding to each candidate surgical path based on the lesion image includes: For each candidate surgical path, based on multiple functional tissue areas corresponding to the target site and the lesion image, the number of target functional tissue areas that the candidate surgical path passes through in the lesion image and the number of first passing pixels corresponding to each target functional tissue area are determined. Based on the target functional organization area and the correspondence between the functional organization area and the weight value, a target weight value corresponding to each target functional organization area is determined; Based on the first number of traversed pixels and the target weight value, determine the functional area feature value corresponding to each target functional organization area; Based on the functional area feature values ​​corresponding to each target functional tissue area, the first path feature information corresponding to each candidate surgical path is determined.

2. The method according to claim 1, characterized in that, The step of determining multiple potential surgical starting points along the edge contour line of the lesion image includes: On the edge contour line of the lesion image, a candidate surgical starting point is determined at every preset number of pixels; or, Based on the length and preset number of edge contour lines, the spacing information between two adjacent candidate surgical starting points is determined, and multiple candidate surgical starting points are determined on the edge contour lines of the lesion image based on the spacing information.

3. The method according to claim 1, characterized in that, If the lesion is located in the head, the corresponding functional tissue areas include the epidermal-cranial region and the brain functional classification region; if the lesion is located in the abdomen, the corresponding functional tissue areas include the skin, connective tissue, and organ classification regions.

4. The method according to claim 3, characterized in that, The weight value corresponding to the epidermal skull region is less than the weight value corresponding to the brain function classification region; the weight value corresponding to the skin is less than the weight value corresponding to the connective tissue, and the weight value corresponding to the connective tissue is less than the weight value corresponding to the organ classification region.

5. The method according to claim 1, characterized in that, The process of determining the target surgical path corresponding to the lesion site from multiple candidate surgical paths, based on a pre-trained target neural network model and the first and second path feature information corresponding to the candidate surgical paths, includes: The first path feature information and the second path feature information corresponding to each candidate surgical path are input into the pre-trained target neural network model to perform path prediction and obtain the prediction probability value corresponding to each candidate surgical path. Based on the predicted probability value, the target surgical path corresponding to the lesion site is determined from multiple candidate surgical paths.

6. A surgical path determination device, characterized in that, include: The image data acquisition module is used to acquire lesion images containing the same lesion site and angiography images corresponding to the lesion images; The candidate path determination module is used to determine multiple candidate surgical starting points on the edge contour line of the lesion image, and to determine multiple candidate surgical paths based on the multiple candidate surgical starting points and the surgical ending points corresponding to the lesion sites. The feature information determination module is used to determine the first path feature information corresponding to each candidate surgical path based on the lesion image, and to determine the second path feature information corresponding to each candidate surgical path based on the angiography image. The target path determination module is used to determine the target surgical path corresponding to the lesion site from multiple candidate surgical paths based on a pre-trained target neural network model, the first path feature information and the second path feature information corresponding to the candidate surgical paths; The edge contour line is the line formed by the edge contour pixels in the lesion image; The feature information determination module includes: a first feature determination unit and a second feature determination unit; The second feature determination unit includes: The pixel determination subunit is used to determine each second crossing pixel in the angiography image through which the candidate surgical path crosses, and to determine the target blood vessel pixel closest to each second crossing pixel. The distance information determination subunit is used to determine the blood vessel distance information corresponding to each second passing pixel based on the second passing pixel and the target blood vessel pixel closest to the second passing pixel; The second feature determination subunit is used to determine the second path feature information corresponding to each candidate surgical path based on the distance information of each blood vessel. The first feature determination unit includes: The number of passes determined subunit is used to determine, for each candidate surgical path, the number of target functional tissue areas and the number of first passing pixels corresponding to each target functional tissue area in the lesion image that the candidate surgical path passes through, based on multiple functional tissue areas and lesion images corresponding to the target site. The target weight value determination sub-unit is used to determine the target weight value corresponding to each target functional organization area based on the target functional organization area and the correspondence between the functional organization area and the weight value. The feature value determination subunit is used to determine the functional area feature value corresponding to each target functional organization area based on the number of first traversed pixels and the target weight value; The first feature determination subunit is used to determine the first path feature information corresponding to each candidate surgical path based on the functional area feature values ​​corresponding to each target functional tissue area.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, which enables the at least one processor to perform the method for determining the surgical path according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for determining the surgical path as described in any one of claims 1-5.

Images