Melanoma lesion position deviation judgment method and system

Abstract

The invention relates to the technical field of medical image processing, and discloses a melanoma lesion position offset judgment method and system.The method comprises the steps that blood vessels and melanoma on a first skin picture are extracted to determine a first positioning datum point, determining the original position of the melanoma in the first skin picture according to the first positioning datum point; after a preset time interval, blood vessels and melanoma on the second skin picture are extracted to determine a second positioning datum point, and the melanoma change position of the melanoma in the second skin picture is determined according to the second positioning datum point; the distance between the original position of the melanoma and the change position of the melanoma can be determined in the first picture; based on the distance between the original position of the melanoma and the change position of the melanoma, determining the position offset of the nidus of the melanoma; the problem that an existing melanoma focus position deviation judgment method is low in judgment precision is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a method and system for determining the position deviation of a melanoma lesion. Background Art

[0002] Melanoma is a highly malignant tumor derived from melanoma, which often occurs in the skin, and can also be seen in mucous membranes and internal organs, and skin malignant melanoma ranks third among skin malignant tumors. Since melanoma often occurs in the human epidermis and has certain color and morphological characteristics, the method of using machine learning technology for early diagnosis of melanoma has received extensive attention from scholars at home and abroad. For example, the invention application CN 108830853 A discloses an artificial intelligence-based melanoma auxiliary diagnosis method, that is, a tumor image on the skin surface is extracted by a machine learning method and then it is judged whether it is a melanoma. However, its judgment is only based on a static method, and in fact, melanoma is gradually developed from melanoma, and its size, shape and position changes within a time scale are related to whether it eventually develops into melanoma. Among them, the size and shape can be obtained by image segmentation after taking a photo, but because people are fat and thin at different times, it is impossible to accurately identify whether their position has changed, which leads to the current low accuracy of judging melanoma. It can be seen that the existing melanoma lesion position offset judgment method has the problem of low judgment accuracy. Summary of the invention

[0003] The present invention provides a method and system for determining the position deviation of a melanoma lesion, so as to solve the problem of low determination accuracy in the existing method for determining the position deviation of a melanoma lesion.

[0004] In order to achieve the above object, the present invention is implemented by the following technical solutions: In a first aspect, the present invention provides a method for determining the position deviation of a melanoma lesion, comprising: Acquire a first skin image containing a melanoma lesion, and extract blood vessels and melanoma on the first skin image; Determine a first positioning reference point based on the blood vessels in the first skin image, and determine the original position of the melanoma in the first skin image according to the first positioning reference point; After a preset time interval, collecting a second skin image including the melanoma lesion in the first skin image, and extracting blood vessels and melanoma on the second skin image; Determine a second positioning reference point based on the blood vessels in the second skin image, and determine a melanoma change position of the melanoma in the second skin image according to the second positioning reference point; Mapping the changed position of the melanoma to the first skin image based on the first positioning reference point and the second positioning reference point, and determining the distance between the original position of the melanoma and the changed position of the melanoma in the first image; The melanoma lesion position deviation is determined based on the distance between the original position of the melanoma and the changed position of the melanoma.

[0005] Optional, such as Figure 2 As shown, extract blood vessels on the skin image, including: Using infrared light to illuminate the skin containing the melanoma lesion and collecting digital image information reflected by the infrared light as a skin picture; The RGB channels of the skin image are enhanced to obtain an enhanced image, and the enhanced image is subjected to denoising preprocessing to extract the blood vessels on the skin image.

[0006] Optionally, extract melanoma on skin images, including: Construct ResNet101 neural network model and Vit-Transformer neural network model, and obtain training data set and validation data set about melanoma through manual annotation; The ResNet101 neural network model and the Vit-Transformer neural network model are combined to construct a melanoma recognition model, and the constructed melanoma recognition model is trained using a training data set and a validation data set to obtain a trained melanoma recognition model; The skin image is input into the trained melanoma recognition model, and the trained melanoma recognition model is used to extract melanoma in the skin image.

[0007] Optionally, determining the first positioning reference point based on the blood vessels in the first skin image includes: Determine the position of each blood vessel in the first skin image, determine the blood vessels that intersect according to the position of each blood vessel, and use the intersection of the blood vessels that intersect as the first positioning reference point, wherein the first positioning reference points in the first skin image include at least three; The determining of the second positioning reference point based on the blood vessels in the second skin image includes: The position of each blood vessel is determined in the second skin image, intersecting blood vessels are determined according to the position of each blood vessel, and the intersection points of the intersecting blood vessels are used as second positioning reference points, wherein the second positioning reference points in the second skin image include at least three.

[0008] Optionally, determining the original position of the melanoma in the first skin image according to the first positioning reference point includes: constructing a first plane coordinate system using the first positioning reference point, constructing a first position mapping relationship between the first skin image and the first plane coordinate system, and determining the position of the melanoma in the first skin image; Mapping the position of the melanoma in the first skin image to the first plane coordinate system using a first mapping relationship to obtain the position of the melanoma in the first plane coordinate system, and using the position of the melanoma in the first plane coordinate system as the original position of the melanoma; The step of determining the changed position of the melanoma in the second skin image according to the second positioning reference point includes: constructing a second plane coordinate system using the second positioning reference point, constructing a second position mapping relationship between the second skin image and the second plane coordinate system, and determining the position of the melanoma in the second skin image; The position of the melanoma in the second skin image is mapped to the second plane coordinate system using a second mapping relationship to obtain the position of the melanoma in the second plane coordinate system, and the position of the melanoma in the second plane coordinate system is used as the melanoma change position.

[0009] Optionally, mapping the melanoma change position to the first skin image based on the first positioning reference point and the second positioning reference point includes: Determine whether the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are the intersection of the same blood vessel. If not, reconstruct the first plane coordinate system and / or the second plane coordinate system until the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are the intersection of the same blood vessel, and re-determine the original position of the melanoma and the changed position of the melanoma. If so, map the position of the changed position of the melanoma in the second plane coordinate system to the first plane coordinate system to obtain the position of the changed position of the melanoma in the first plane coordinate system.

[0010] Optionally, the method for determining whether the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are intersection points of the same blood vessel includes: Extracting first shape information of the blood vessel corresponding to a first positioning reference point used to construct a first plane coordinate system, and extracting second shape information of the blood vessel corresponding to a second positioning reference point used to construct a second plane coordinate system; Comparing the first shape information with the second shape information to determine whether the first shape information and the second shape information are consistent, and determining that a first positioning reference point for constructing the first plane coordinate system and a second positioning reference point for constructing the second plane coordinate system are the same blood vessel if it is determined that they are consistent, and determining that the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are different blood vessels if it is determined that they are not consistent; The process of comparing the first shape information with the second shape information includes enlarging, reducing, and rotating the first shape information and / or the second shape information.

[0011] Optionally, determining the distance between the original position of the melanoma and the changed position of the melanoma in the first image includes: The position of the changed position of the melanoma in the first plane coordinate system is compared with the original position of the melanoma to obtain first distance information, and the first distance information is used as the distance between the original position of the melanoma and the changed position of the melanoma.

[0012] Optionally, determining the melanoma lesion position shift based on the distance between the original position of the melanoma and the changed position of the melanoma includes: The first distance information is compared with a preset first threshold value, and when the first distance information is less than or equal to the preset first threshold value, it is judged that the position of the melanoma lesion has not shifted; when the first distance information is greater than the preset first threshold value, the second distance information of the original position of the melanoma and all the first positioning reference points is determined; at the same time, based on the position of the changed position of the melanoma in the first plane coordinate system, the third distance information of the changed position of the melanoma and all the first positioning reference points is determined, and the difference between the second distance information and the third distance information is calculated; when the difference is less than or equal to the preset second threshold value, it is judged that the position of the melanoma lesion has not shifted; when the difference is greater than the preset second threshold value, it is judged that the position of the melanoma lesion has shifted.

[0013] In a second aspect, an embodiment of the present application provides a melanoma lesion position deviation determination system, including a processor and a memory; Memory, used to store computer programs; The processor is used to implement any method step described in the first aspect when executing a program stored in the memory.

[0014] Beneficial effects: The method for determining the position deviation of a melanoma lesion provided by the present invention determines a first positioning reference point by extracting blood vessels and melanoma on a first skin image, and determines the original position of the melanoma in the first skin image according to the first positioning reference point; extracts blood vessels and melanoma on a second skin image after a preset time interval to determine a second positioning reference point, and determines the changed position of the melanoma in the second skin image according to the second positioning reference point; the distance between the original position of the melanoma and the changed position of the melanoma can be determined in the first image; thereby locating the position of the melanoma, and then determining whether it has position deviation. Since blood vessels are usually stationary, using blood vessels as a reference greatly improves the accuracy of melanoma diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a method for determining the position deviation of a melanoma lesion according to a preferred embodiment of the present invention; Figure 2 A schematic diagram of extracting blood vessels from a skin image according to a preferred embodiment of the present invention; Figure 3 Schematic diagram of the structure of a melanoma recognition model according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0016] The technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not indicate quantity restrictions, but indicate the existence of at least one. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0018] See also Figure 1 The present application provides a method for determining the position deviation of a melanoma lesion, comprising: Acquire a first skin image containing a melanoma lesion, and extract blood vessels and melanoma on the first skin image; Determine a first positioning reference point based on the blood vessels in the first skin image, and determine the original position of the melanoma in the first skin image according to the first positioning reference point; After a preset time interval, collecting a second skin image including the melanoma lesion in the first skin image, and extracting blood vessels and melanoma on the second skin image; Determine a second positioning reference point based on the blood vessels in the second skin image, and determine a melanoma change position of the melanoma in the second skin image according to the second positioning reference point; Mapping the changed position of the melanoma to the first skin image based on the first positioning reference point and the second positioning reference point, and determining the distance between the original position of the melanoma and the changed position of the melanoma in the first image; The melanoma lesion position deviation is determined based on the distance between the original position of the melanoma and the changed position of the melanoma.

[0019] In this embodiment, a first positioning reference point is determined by extracting blood vessels and melanoma on a first skin image, and the original position of the melanoma in the first skin image is determined based on the first positioning reference point; after a preset time interval, blood vessels and melanoma are extracted on a second skin image to determine a second positioning reference point, and the changed position of the melanoma in the second skin image is determined based on the second positioning reference point; the distance between the original position of the melanoma and the changed position of the melanoma can be determined in the first image; thereby, the position of the melanoma is located, and then it is determined whether it has a positional offset. Since blood vessels are usually stationary, using blood vessels as a reference greatly improves the accuracy of melanoma diagnosis.

[0020] The preset time interval in the embodiment can be divided into two situations: 1. It is recommended to conduct a clinical evaluation every 3 months after surgery; 2. For patients who have completed the initial treatment, the recommended follow-up plan is every 3 months in the first year, every 4 months in the second year, every 6 months from the third to the fifth year, and once a year after five years. Therefore, according to the first situation, the preset time interval can be set to 3 months, and according to the second situation, the preset time interval can be set to 3 months in the first year, 4 months in the second year, 6 months from the third to the fifth year, and 12 months after five years.

[0021] Regarding the selection of the melanoma position in the embodiment, no matter it is the original position of the melanoma or the changed position of the melanoma, the centroid position of the corresponding melanoma can be selected as the melanoma position to facilitate the calculation.

[0022] Regarding the selection of positioning reference points, since the position of blood vessels will basically not change without the influence of external forces, the position of blood vessels can be used as the selection standard of reference points. The intersection between blood vessels is the natural positioning point. Therefore, the intersection of blood vessels on the skin image can be used as the positioning reference point to determine the position of the skin image.

[0023] Optionally, extract blood vessels from the skin image, including: like Figure 2 As shown, infrared light is used to irradiate skin containing melanoma lesions, and digital image information reflected by the infrared light is collected as a skin picture; The RGB channels of the skin image are enhanced to obtain an enhanced image, and the enhanced image is subjected to denoising preprocessing to extract the blood vessels on the skin image.

[0024] Optionally, extract melanoma on skin images, including: Construct ResNet101 neural network model and Vit-Transformer neural network model, and obtain training data set and validation data set about melanoma through manual annotation; The ResNet101 neural network model and the Vit-Transformer neural network model are combined to construct a melanoma recognition model, and the constructed melanoma recognition model is trained using a training data set and a validation data set to obtain a trained melanoma recognition model; The skin image is input into the trained melanoma recognition model, and the trained melanoma recognition model is used to extract melanoma in the skin image.

[0025] In this embodiment, ResNet-101 is a deep convolutional neural network, whose main feature is the use of residual blocks to solve the gradient vanishing problem in deep network training. In deep learning, as the number of network layers increases, the gradient vanishing problem becomes more and more serious, making it difficult for the network to converge. ResNet-101 uses residual blocks to build a deep network, making it easier to train the network and achieving excellent results in image recognition tasks.

[0026] The mathematical expression of the ResNet-101 model is as follows: Input: X Residual block: F(X)+X Among them, F(X) represents the mapping function of the residual block, and X represents the input data. The residual block can be regarded as a skip connection, which allows the input data to be directly passed to the output layer, thus avoiding the influence of the gradient vanishing problem.

[0027] There are 101 layers in the ResNet101 network structure diagram, the first layer is a 7x7 convolutional layer, followed by 4 stages, each stage contains several residual blocks.

[0028] Finally, there is a global average pooling layer and a fully connected layer.

[0029] Each residual block consists of two 3x3 convolutional layers, each followed by a batch normalization and ReLU activation function. There is also batch normalization and ReLU activation function between residual blocks, but no convolutional layers. The first residual block in each stage uses a 1x1 convolutional layer to convert the number of input channels to the number of output channels for summing with subsequent residual blocks.

[0030] Vision Transformer (ViT) is a deep learning model based on the Transformer architecture for image recognition and computer vision tasks. Unlike traditional convolutional neural networks (CNNs), ViT directly treats images as a sequential input and uses a self-attention mechanism to process pixel relationships in images.

[0031] ViT works by dividing the image into a series of patches and converting each patch into a vector representation as an input sequence. These vectors are then processed through a multi-layer Transformer encoder, which includes a self-attention mechanism and a feed-forward neural network layer. This captures the contextual dependencies between different locations in the image. Finally, by classifying or regressing the Transformer encoder output, specific visual tasks can be completed.

[0032] like Figure 3 As shown in the figure, vit is the image block applied to the transformer. CNN uses the convolution kernel to convolve the image based on the sliding window idea to obtain the feature map. In order to make the image imitate the input sequence of NLP, we can first divide the image into blocks (patches), then flatten these image blocks and input them into the network (so that it becomes an image sequence), then extract features through the transformer, and finally classify these features through MLP. In fact, it can be understood as replacing the backbone with the transformer in the previous CNN classification task.

[0033] Optionally, determining the first positioning reference point based on the blood vessels in the first skin image includes: Determine the position of each blood vessel in the first skin image, determine the blood vessels that intersect according to the position of each blood vessel, and use the intersection of the blood vessels that intersect as the first positioning reference point, wherein the first positioning reference points in the first skin image include at least three; The determining of the second positioning reference point based on the blood vessels in the second skin image includes: The position of each blood vessel is determined in the second skin image, intersecting blood vessels are determined according to the position of each blood vessel, and the intersection points of the intersecting blood vessels are used as second positioning reference points, wherein the second positioning reference points in the second skin image include at least three.

[0034] Optionally, determining the original position of the melanoma in the first skin image according to the first positioning reference point includes: constructing a first plane coordinate system using the first positioning reference point, constructing a first position mapping relationship between the first skin image and the first plane coordinate system, and determining the position of the melanoma in the first skin image; Mapping the position of the melanoma in the first skin image to the first plane coordinate system using a first mapping relationship to obtain the position of the melanoma in the first plane coordinate system, and using the position of the melanoma in the first plane coordinate system as the original position of the melanoma; The step of determining the changed position of the melanoma in the second skin image according to the second positioning reference point includes: constructing a second plane coordinate system using the second positioning reference point, constructing a second position mapping relationship between the second skin image and the second plane coordinate system, and determining the position of the melanoma in the second skin image; The position of the melanoma in the second skin image is mapped to the second plane coordinate system using a second mapping relationship to obtain the position of the melanoma in the second plane coordinate system, and the position of the melanoma in the second plane coordinate system is used as the melanoma change position.

[0035] Optionally, mapping the melanoma change position to the first skin image based on the first positioning reference point and the second positioning reference point includes: Determine whether the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are the intersection of the same blood vessel. If not, reconstruct the first plane coordinate system and / or the second plane coordinate system until the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are the intersection of the same blood vessel, and re-determine the original position of the melanoma and the changed position of the melanoma. If so, map the position of the changed position of the melanoma in the second plane coordinate system to the first plane coordinate system to obtain the position of the changed position of the melanoma in the first plane coordinate system.

[0036] Optionally, the method for determining whether the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are intersection points of the same blood vessel includes: Extracting first shape information of the blood vessel corresponding to a first positioning reference point used to construct a first plane coordinate system, and extracting second shape information of the blood vessel corresponding to a second positioning reference point used to construct a second plane coordinate system; Comparing the first shape information with the second shape information to determine whether the first shape information and the second shape information are consistent, and determining that a first positioning reference point for constructing the first plane coordinate system and a second positioning reference point for constructing the second plane coordinate system are the same blood vessel if it is determined that they are consistent, and determining that the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are different blood vessels if it is determined that they are not consistent; The process of comparing the first shape information with the second shape information includes enlarging, reducing, and rotating the first shape information and / or the second shape information.

[0037] Optionally, determining the distance between the original position of the melanoma and the changed position of the melanoma in the first image includes: The position of the changed position of the melanoma in the first plane coordinate system is compared with the original position of the melanoma to obtain first distance information, and the first distance information is used as the distance between the original position of the melanoma and the changed position of the melanoma.

[0038] Optionally, determining the melanoma lesion position shift based on the distance between the original position of the melanoma and the changed position of the melanoma includes: The first distance information is compared with a preset first threshold value, and when the first distance information is less than or equal to the preset first threshold value, it is judged that the position of the melanoma lesion has not shifted; when the first distance information is greater than the preset first threshold value, the second distance information of the original position of the melanoma and all the first positioning reference points is determined; at the same time, based on the position of the changed position of the melanoma in the first plane coordinate system, the third distance information of the changed position of the melanoma and all the first positioning reference points is determined, and the difference between the second distance information and the third distance information is calculated; when the difference is less than or equal to the preset second threshold value, it is judged that the position of the melanoma lesion has not shifted; when the difference is greater than the preset second threshold value, it is judged that the position of the melanoma lesion has shifted.

[0039] The embodiment of the present application also provides a melanoma lesion position deviation judgment system, including a processor and a memory; Memory, used to store computer programs; The processor is used to implement any method step in the method for determining the position deviation of a melanoma lesion when executing the program stored in the memory.

[0040] The above-mentioned melanoma lesion position deviation judgment system can implement various embodiments of the above-mentioned melanoma lesion position deviation judgment method and can achieve the same beneficial effects, which will not be described in detail here.

[0041] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for determining the position deviation of a melanoma lesion, characterized in that: include: Acquire a first skin image containing a melanoma lesion, and extract blood vessels and melanoma on the first skin image; Determine a first positioning reference point based on the blood vessels in the first skin image, and determine the original position of the melanoma in the first skin image according to the first positioning reference point; After a preset time interval, collecting a second skin image including the melanoma lesion in the first skin image, and extracting blood vessels and melanoma on the second skin image; Determine a second positioning reference point based on the blood vessels in the second skin image, and determine a melanoma change position of the melanoma in the second skin image according to the second positioning reference point; Mapping the changed position of the melanoma to the first skin image based on the first positioning reference point and the second positioning reference point, and determining the distance between the original position of the melanoma and the changed position of the melanoma in the first image; The melanoma lesion position deviation is determined based on the distance between the original position of the melanoma and the changed position of the melanoma.

2. The method for determining the position deviation of a melanoma lesion according to claim 1, characterized in that: Extract blood vessels from skin images, including: Using infrared light to illuminate the skin containing the melanoma lesion and collecting digital image information reflected by the infrared light as a skin picture; The RGB channels of the skin image are enhanced to obtain an enhanced image, and the enhanced image is subjected to denoising preprocessing to extract the blood vessels on the skin image.

3. The method for determining the position deviation of a melanoma lesion according to claim 1, characterized in that: Extract melanoma on skin images, including: Construct ResNet101 neural network model and Vit-Transformer neural network model, and obtain training data set and validation data set for melanoma through manual annotation; The ResNet101 neural network model and the Vit-Transformer neural network model are combined to construct a melanoma recognition model, and the constructed melanoma recognition model is trained using a training data set and a validation data set to obtain a trained melanoma recognition model; The skin image is input into the trained melanoma recognition model, and the trained melanoma recognition model is used to extract melanoma in the skin image.

4. The method for determining the position deviation of a melanoma lesion according to claim 1, characterized in that: The step of determining the first positioning reference point based on the blood vessels in the first skin image includes: Determine the position of each blood vessel in the first skin image, determine the blood vessels that intersect according to the position of each blood vessel, and use the intersection of the blood vessels that intersect as the first positioning reference point, wherein the first positioning reference points in the first skin image include at least three; The determining of the second positioning reference point based on the blood vessels in the second skin image includes: The position of each blood vessel is determined in the second skin image, intersecting blood vessels are determined according to the position of each blood vessel, and the intersection points of the intersecting blood vessels are used as second positioning reference points, wherein the second positioning reference points in the second skin image include at least three.

5. The method for determining the position deviation of a melanoma lesion according to claim 4, characterized in that: Determining the original position of the melanoma in the first skin image according to the first positioning reference point includes: constructing a first plane coordinate system using the first positioning reference point, constructing a first position mapping relationship between the first skin image and the first plane coordinate system, and determining the position of the melanoma in the first skin image; Mapping the position of the melanoma in the first skin image to the first plane coordinate system using a first mapping relationship to obtain the position of the melanoma in the first plane coordinate system, and using the position of the melanoma in the first plane coordinate system as the original position of the melanoma; The step of determining the changed position of the melanoma in the second skin image according to the second positioning reference point includes: constructing a second plane coordinate system using the second positioning reference point, constructing a second position mapping relationship between the second skin image and the second plane coordinate system, and determining the position of the melanoma in the second skin image; The position of the melanoma in the second skin image is mapped to the second plane coordinate system using a second mapping relationship to obtain the position of the melanoma in the second plane coordinate system, and the position of the melanoma in the second plane coordinate system is used as the melanoma change position.

6. The method for determining the position deviation of a melanoma lesion according to claim 5, characterized in that: Mapping the melanoma change position to the first skin image based on the first positioning reference point and the second positioning reference point includes: Determine whether the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are the intersection of the same blood vessel. If not, reconstruct the first plane coordinate system and / or the second plane coordinate system until the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are the intersection of the same blood vessel, and re-determine the original position of the melanoma and the changed position of the melanoma. If so, map the position of the changed position of the melanoma in the second plane coordinate system to the first plane coordinate system to obtain the position of the changed position of the melanoma in the first plane coordinate system.

7. The method for determining the position deviation of a melanoma lesion according to claim 6, characterized in that: The method for determining whether a first positioning reference point for constructing a first plane coordinate system and a second positioning reference point for constructing a second plane coordinate system are intersection points of the same blood vessel includes: Extracting first shape information of the blood vessel corresponding to a first positioning reference point used to construct a first plane coordinate system, and extracting second shape information of the blood vessel corresponding to a second positioning reference point used to construct a second plane coordinate system; Comparing the first shape information with the second shape information to determine whether the first shape information and the second shape information are consistent, and determining that a first positioning reference point for constructing the first plane coordinate system and a second positioning reference point for constructing the second plane coordinate system are the same blood vessel if it is determined that they are consistent, and determining that the first positioning reference point for constructing the first plane coordinate system and the second positioning reference point for constructing the second plane coordinate system are different blood vessels if it is determined that they are not consistent; The process of comparing the first shape information with the second shape information includes enlarging, reducing, and rotating the first shape information and / or the second shape information.

8. The method for determining the position deviation of a melanoma lesion according to claim 7, characterized in that: Determining the distance between the original position of the melanoma and the changed position of the melanoma in the first image includes: The position of the changed position of the melanoma in the first plane coordinate system is compared with the original position of the melanoma to obtain first distance information, and the first distance information is used as the distance between the original position of the melanoma and the changed position of the melanoma.

9. The method for determining the position deviation of a melanoma lesion according to claim 8, characterized in that: Determining the displacement of the melanoma lesion position based on the distance between the original position of the melanoma and the changed position of the melanoma includes: The first distance information is compared with a preset first threshold value, and when the first distance information is less than or equal to the preset first threshold value, it is judged that the position of the melanoma lesion has not shifted; when the first distance information is greater than the preset first threshold value, the second distance information of the original position of the melanoma and all the first positioning reference points is determined; at the same time, based on the position of the changed position of the melanoma in the first plane coordinate system, the third distance information of the changed position of the melanoma and all the first positioning reference points is determined, and the difference between the second distance information and the third distance information is calculated; when the difference is less than or equal to the preset second threshold value, it is judged that the position of the melanoma lesion has not shifted; when the difference is greater than the preset second threshold value, it is judged that the position of the melanoma lesion has shifted.

10. A melanoma lesion position deviation judgment system, characterized in that: Including processor and memory; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 9 when executing a program stored in a memory.

Images