Acupuncture point identification method, system and equipment for acupuncture part and medium

Abstract

The invention belongs to the technical field of acupuncture and moxibustion, and provides an acupuncture point recognition method, system and device for an acupuncture part and a medium, the method comprises the following steps: establishing a body surface anatomical marker coordinate system, and determining body surface coordinates of human body acupuncture points in the body surface anatomical marker coordinate system; determining coordinates of an acupuncture part in the body surface anatomical marker coordinate system; human body acupuncture point body surface coordinates near the acupuncture part coordinates are determined; calculating the distance between the coordinates of the acupuncture part and the body surface coordinates of the acupuncture points of the human body near the acupuncture part; determining an acupuncture point of the acupuncture part according to the distance and a preset acupuncture point weight; by calculating the spatial distance between the recognized coordinates of the acupuncture part and the body surface coordinates of the human body acupuncture points near the body surface of the patient and weighting the scores of the corresponding acupuncture points according to the acupuncture point use frequency, the acupuncture points of the acupuncture part are determined according to the scores, clinical experience factors are added for acupuncture point recognition, and the recognition accuracy is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of acupuncture, and in particular relates to a method, system, equipment and medium for identifying acupuncture points. Background Art

[0002] Acupuncture, as a treasure of traditional Chinese medicine, has been recognized by the World Health Organization (WHO) as an appropriate method for treating a variety of diseases. Although acupuncture is widely used in clinical treatment, the lack of clinical acupoint records has hindered the standardization of acupuncture medical records and the development of acupuncture decision-making support systems. Especially in the era of big data, the lack of acupoint prescription records for acupuncture has led to the lack of key underlying data for the intelligentization of acupuncture big data. The traditional manual method of recording acupoints is time-consuming and inefficient, which makes it difficult for clinicians to record the names of acupoints used in real time.

[0003] With the development of computer vision and deep learning technology, many methods of identifying acupoints through neural networks have emerged in the existing technology. However, some acupoints are too close to each other and are difficult to distinguish by visual recognition and distance calculation alone, which affects the accuracy of acupoint recognition. Summary of the invention

[0004] In view of the problems in the background technology, the present invention proposes a method, system, device and medium for identifying acupuncture points.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present disclosure provides a method for identifying acupuncture points, comprising: Establishing a body surface anatomical landmark coordinate system, and determining the body surface coordinates of human acupuncture points in the body surface anatomical landmark coordinate system; Determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system; Determine the body surface coordinates of the acupuncture points near the acupuncture site coordinates; Calculating the distance between the coordinates of the acupuncture site and the coordinates of the human body acupuncture points near the acupuncture site; The acupoints at the acupuncture site are determined according to the distance and the preset acupoint weights.

[0006] Preferably, the establishing of the body surface anatomical landmark coordinate system comprises: Acquire a human body surface image by using an image acquisition device; Preprocessing the human body surface image; Inputting the preprocessed human body surface image into the trained first OpenPose model to obtain the coordinates of key points of the human body; A body surface anatomical landmark coordinate system is established according to the coordinates of the key points of the human body and the predetermined body surface anatomical landmark positions and the human body coordinate system.

[0007] Preferably, the preprocessing of the human body surface image includes: Converting the color space of the human body surface image to HSV; The human body contour in the human body surface image is determined by the YOLOv7 model and the Mask R-CNN model.

[0008] Preferably, the determining of the body surface coordinates of the human acupuncture points in the body surface anatomical landmark coordinate system comprises: Obtaining the surface anatomical landmarks corresponding to the target acupuncture points; The body surface coordinates of the human acupuncture points are determined based on the bone measurement method and through the body surface anatomical landmarks corresponding to the target acupuncture points.

[0009] Preferably, the determining of the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system comprises: Acquire acupuncture action video through image acquisition equipment; Inputting the acupuncture action video into a trained acupuncture gesture recognition model to obtain acupuncture action occurrence frames; According to the acupuncture action occurrence frame, the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system are determined.

[0010] Preferably, training the acupuncture gesture recognition model includes: Get the acupuncture action training video; Preprocessing the acupuncture action training video; Extracting images when acupuncture action occurs and images when no acupuncture action occurs from the acupuncture action training video, and manually labeling them to obtain the category of each acupuncture action; Using the trained second OpenPose model to extract the image when the acupuncture action occurs, and obtain the coordinates of the key points of the hand; Calculating the characteristic vector of the acupuncture action according to the coordinates of the key points of the hand; According to the coordinates of the key points of the hand and the feature vectors of the acupuncture action, the data annotates the characteristics of each acupuncture action; Based on the images when the acupuncture action occurs and the images when no acupuncture action occurs, as well as the category and features of each acupuncture action, a labeled acupuncture action dataset is constructed; Based on the acupuncture action data set, an acupuncture gesture recognition model is trained to obtain a trained acupuncture gesture recognition model.

[0011] Preferably, the step of determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system according to the acupuncture action occurrence frame includes: Determine the three-dimensional coordinates of the physician's fingertip in the acupuncture action occurrence frame in the body surface anatomical landmark coordinate system; The three-dimensional coordinates of the doctor's fingertip in the body surface anatomical landmark coordinate system are used as the acupuncture site coordinates.

[0012] In a second aspect, the present disclosure provides a system for identifying acupuncture points, comprising: A body surface coordinate determination module for acupuncture points on the human body, used to establish a body surface anatomical landmark coordinate system and determine the body surface coordinates of acupuncture points on the human body in the body surface anatomical landmark coordinate system; A module for determining the coordinates of the acupuncture site, used for determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system; A neighboring acupuncture point determination module is used to determine the body surface coordinates of human acupuncture points near the coordinates of the acupuncture site; A distance calculation module, used to calculate the distance between the coordinates of the acupuncture site and the coordinates of the human body acupoints near the acupuncture site; The acupuncture site acupoint determination module is used to determine the acupoints at the acupuncture site according to the distance and the preset acupoint weights.

[0013] In a third aspect, the present disclosure proposes a device comprising a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, and when the processor executes the computer instructions, the method for identifying acupuncture points described in the first aspect is executed.

[0014] In a fourth aspect, the present disclosure proposes a computer-readable storage medium having computer instructions stored thereon. When the computer instructions are executed, the acupuncture point identification method described in the first aspect can be implemented.

[0015] Beneficial effects of the present invention: The method of the present invention calculates the spatial distance between the coordinates of the identified acupuncture site and the surface coordinates of the human acupoints near the patient's body surface, and weights the scores of the corresponding acupoints by the frequency of acupoint use, thereby determining the acupoints of the acupuncture site according to the scores, adding clinical experience factors to acupuncture point identification, and improving identification accuracy.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become obvious from the description, or be understood by implementing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flow chart showing a method for identifying acupuncture points of the present invention is shown; Figure 2 A framework diagram of an acupuncture point identification system of the present invention is shown; Figure 3 A schematic diagram of the device structure of the present invention is shown. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0020] Reference Figure 1 As shown, a method for identifying acupuncture points comprises the following steps: S1. Establishing a body surface anatomical landmark coordinate system, and determining the body surface coordinates of human acupuncture points in the body surface anatomical landmark coordinate system; Specifically, a method for establishing a body surface anatomical landmark coordinate system and determining the body surface coordinates of human acupuncture points in the body surface anatomical landmark coordinate system comprises the following steps: S11, acquiring a human body surface image through an image acquisition device; Among them, the image acquisition device is a depth camera. By shooting the human body surface image with a depth camera, not only can a two-dimensional color image (RGB image) be captured, but also the depth information of each pixel can be recorded at the same time. Therefore, each pixel has not only color information but also depth information. Therefore, by analyzing the two-dimensional coordinates, depth value and camera focal length of each pixel in the human body surface image, a three-dimensional spatial coordinate can be established for each pixel. Specifically, a video containing the patient's body is shot with a depth camera, and then the human body surface image in the video is extracted. When shooting the video, the patient needs to remain in a supine or prone position and ensure that the body surface is fully exposed.

[0021] S12, preprocessing human body surface images; Specifically, the method for preprocessing a human body surface image comprises the following steps: S121, converting the color space of the human body surface image into HSV; Specifically, by converting the human body surface image from the RGB color space to the HSV color space, the skin and the background can be effectively distinguished, which can improve the accuracy of human skin color detection.

[0022] S122, determine the human body contour in the human body surface image by using the YOLOv7 model and the Mask R-CNN model; Specifically, the YOLOv7 model is used to detect the human body contour in the human body surface image, and the Mask R-CNN model is used to perform instance segmentation on the human body surface image to identify the occluded human body parts, which reduces the amount of calculation for the subsequent recognition of the surface anatomical landmarks. It should be further explained that the YOLOv7 model can efficiently detect the human body contour in the image, which is a mature application in the prior art. Therefore, this method directly uses the trained YOLOv7 model to detect the human body contour in the human body surface image. For those skilled in the art, how to train the YOLOv7 model and the specific detection process are common knowledge in the art and no further explanation is required. Similarly, the Mask R-CNN model can efficiently perform instance segmentation, accurately identify and segment the occluded human body parts in the image, which is a mature application in the prior art. Therefore, this method directly uses the trained Mask R-CNN model to identify and segment the occluded human body parts in the image. For those skilled in the art, how to train the Mask R-CNN model and the specific recognition process are common knowledge in the art and no further explanation is required.

[0023] S13, inputting the preprocessed human body surface image into the trained first OpenPose model to obtain the coordinates of key points of the human body; The OpenPose model is a model that has been trained with a large-scale universal human posture dataset and can detect key points of the human body in an image (such as shoulders, elbows, knees, hands, etc.). The method for training the first OpenPose model is to obtain existing datasets of labeled human body surface markers (Universal Pose-3 dataset and DensePose dataset), then perform data enhancement on the images in the existing datasets of labeled human body surface markers, and finally use the data-enhanced dataset of human body surface markers to fine-tune the first OpenPose model to ensure that the first OpenPose model can accurately identify specific anatomical landmarks in acupuncture treatment.

[0024] S14, establishing a body surface anatomical landmark coordinate system according to the coordinates of the key points of the human body and the predetermined body surface anatomical landmark positions and the human body coordinate system; Specifically, the method for predetermining the human body coordinate system includes the following steps: 1. Select the center of gravity or pelvic center of the human body as the coordinate origin to improve the stability of the coordinate system in different postures.

[0025] 2. Define the coordinate axes according to the front-back, left-right, and up-down directions of the human body, and set the positive direction of the coordinate axes based on human anatomy knowledge to establish the human body coordinate system. For example: the head facing up is the positive direction of the Z axis, the spine facing forward is the positive direction of the Y axis, and the right hand facing right is the positive direction of the X axis.

[0026] Specifically, the method for establishing the coordinates of the body surface anatomical landmarks is as follows: based on the established body surface anatomical landmark coordinate system and the body posture and shooting angle, the body coordinate system is mapped to the real-time changing 3D image coordinate system to establish the body surface anatomical landmark coordinate system.

[0027] In the body surface anatomical landmark coordinate system, the position of each body surface anatomical landmark in the body surface anatomical landmark coordinate system is a three-dimensional coordinate point , which can provide accurate spatial reference for subsequent acupoint positioning.

[0028] S15, determining the body surface coordinates of the human acupuncture points in the body surface anatomical landmark coordinate system; Specifically, in the established surface anatomical landmark coordinate system, the bone measurement method is used to determine the surface coordinates of human acupuncture points. Furthermore, the corresponding surface anatomical landmarks with clear markings of the target acupuncture points to be determined, such as joint bone protrusions, are obtained, and then based on the distance and direction of the bone measurement, starting from the known surface anatomical landmarks, the surface coordinates of the human acupuncture points in the surface anatomical landmark coordinate system are determined according to the proportion and direction of the bone measurement, and this is repeated until the surface coordinates of all human acupuncture points in the surface anatomical landmark coordinate system are determined. Exemplarily, the position of Zusanli is defined as 3 / 16 of the line connecting Dubi point and Jiexi point, where Dubi point is the midpoint of the depression on the outside of the lower edge of the patella, and Jiexi point is the midpoint of the central depression of the transverse line at the junction of the dorsum of the foot and the calf.

[0029] S2. determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system; Specifically, the method for determining the acupuncture site includes the following steps: S21, obtaining acupuncture action video through an image acquisition device; Among them, the acupuncture action video includes the doctor's hands to capture the specific positions of the fingers, so that the acupuncture gesture recognition model can recognize the action frames.

[0030] S22, inputting the acupuncture action video into the trained acupuncture gesture recognition model to obtain the acupuncture action occurrence frame; Specifically, the method for training the acupuncture gesture recognition model includes the following steps: S221. Obtain acupuncture action training video; specifically, the acupuncture action video includes the complete action process of the physician holding the needle, pressing the acupoint, and acupuncture.

[0031] The piercing hand is the hand that the acupuncturist holds the needle during acupuncture. When acupuncture occurs, the piercing hand usually presses the skin of the acupoint area with the middle finger and uses the thumb and index finger to hold the needle body and pierce the skin.

[0032] S222, pre-treatment acupuncture action training video; Specifically, the method of preprocessing the acupuncture action training video includes: removing the background, adjusting the brightness and contrast, etc., to ensure the image quality and consistency; S223, extracting key frames (i.e., images when acupuncture action occurs) and non-key frames (images when no acupuncture action occurs) from the acupuncture action training video, and manually labeling them to obtain the category of each acupuncture action; Manual annotation is to mark each image whether it is an image of acupuncture. Among them, the action of the index finger and thumb pinching the needle and moving it quickly towards the middle finger is acupuncture.

[0033] S224, using the trained second OpenPose model to extract the image when the acupuncture action occurs, and obtain the coordinates of the key points of the hand; The hand key point coordinates include the image three-dimensional coordinates of the hand key points such as the index finger, thumb and middle finger.

[0034] Specifically, the second OpenPose model is trained using the OneHand10K dataset (hand landmark dataset) to make it easier to recognize hand movements. The OneHand10K dataset is a publicly available dataset.

[0035] S225, calculating the characteristic vector of the acupuncture action according to the coordinates of the key points of the hand; Among them, the characteristic vectors of acupuncture action include: the distance between fingers, the distance between fingers and the patient's body surface, the angle of fingers, etc.

[0036] S226, data annotating the features of each acupuncture action according to the coordinates of the key points of the hand and the feature vectors of the acupuncture action; Through data annotation, it is possible to prepare annotated data for acupuncture gesture recognition model training; S227, constructing a labeled acupuncture action dataset based on the images when acupuncture action occurs and the images when no acupuncture action occurs, as well as the category and features of each acupuncture action; S227, training an acupuncture gesture recognition model based on the acupuncture action data set to obtain a trained acupuncture gesture recognition model; Specifically, a convolutional neural network (CNN) model was trained using a labeled acupuncture action dataset. The model was trained by setting the network structure and parameters of CNN, and the model performance was optimized based on the performance of the validation set. Finally, the performance of the trained model was evaluated using the test set. The main evaluation indicators included accuracy, recall, and precision. Finally, a trained CNN model, namely the acupuncture gesture recognition model, was obtained, which was used to classify acupuncture action frames and non-acupuncture action frames.

[0037] S23, determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system according to the acupuncture action occurrence frame; Specifically, firstly, the three-dimensional coordinates of the doctor's fingertips (thumb, index finger, and middle finger tips) in the acupuncture action frame in the patient's body surface anatomical landmark coordinate system are selected, and then the three-dimensional coordinates of the doctor's fingertips in the body surface anatomical landmark coordinate system are used as the coordinates of the acupuncture site.

[0038] S3. Determine the acupuncture points; S31, determining the body surface coordinates of acupuncture points near the coordinates of the acupuncture site; Specifically, the surface distances between the body surface coordinates of all human acupoints and the coordinates of the acupuncture site are first calculated; then, with half the width of the human model's palm as the radius, the acupoints whose surface distances are less than the radius are screened out as the adjacent acupoints of the acupuncture site.

[0039] S32, calculating the distance between the coordinates of the acupuncture site and the coordinates of the human body surface of the acupuncture points near the acupuncture site; Specifically, the coordinates of the acupuncture site can be expressed as: , the body surface coordinates of the acupuncture points near the acupuncture site can be expressed as: , the three-dimensional Euclidean distance formula is used to calculate the distance between the coordinates of the acupuncture site and the coordinates of the human body surface near the acupuncture site , whose expression is: .

[0040] S33, determining the acupoints at the acupuncture site according to the distance and the preset acupoint weights; However, relying solely on distance calculation may result in situations where multiple adjacent acupoints are difficult to distinguish, so acupoint weights need to be introduced to ensure accuracy.

[0041] Specifically, the method of presetting the acupoint weight is: counting the usage frequency information of each acupoint, and introducing an acupoint weight adjustment parameter according to the usage frequency information of each acupoint , which is used to adjust the influence of acupoint usage frequency, and its expression is: , where For acupuncture points Furthermore, the method for counting the frequency of use of each acupoint is to use "acupoint use rules" as the main search term in literature platforms such as HowNet, Wanfang, and PubMed, search for relevant literature, count the frequency of use of each acupoint in different diseases, and perform standardized processing to form a standardized acupoint use frequency pool.

[0042] Specifically, the method for determining the acupuncture points is as follows: taking into account the weight and distance, and calculating the score of each acupuncture point , whose expression is: ,Finally, the scores of each acupoint are sorted and the name of the acupoint with the highest score is output, which is the acupoint at the acupuncture site.

[0043] By calculating the spatial distance between the coordinates of the identified acupuncture site and the surface coordinates of the human acupoints near the patient's body surface, and weighting the scores of the corresponding acupoints by the frequency of acupoint use, the acupoints of the acupuncture site are determined according to the scores, adding clinical experience factors to acupuncture point identification and improving recognition accuracy.

[0044] Based on the same inventive concept, Figure 2 As shown, the present disclosure provides a system for identifying acupuncture points, comprising: A body surface coordinate determination module 110 for acupuncture points on the human body is used to establish a body surface anatomical landmark coordinate system and determine the body surface coordinates of acupuncture points on the human body in the body surface anatomical landmark coordinate system; Acupuncture site coordinate determination module 120, used for determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system; The adjacent acupuncture point determination module 130 is used to determine the body surface coordinates of the human acupuncture points near the coordinates of the acupuncture site; A distance calculation module 140 is used to calculate the distance between the coordinates of the acupuncture site and the coordinates of the human body surface near the acupuncture site; The acupuncture site acupoint determination module 150 is used to determine the acupoints of the acupuncture site according to the distance and the preset acupoint weights.

[0045] Based on the same inventive concept, Figure 3 As shown, the present disclosure proposes a device including a memory and a processor, wherein the memory stores computer instructions that can be run on the processor, and the processor executes the above-mentioned acupuncture point identification method when running the computer instructions.

[0046] Based on the same inventive concept, the present disclosure proposes a computer-readable storage medium on which computer instructions are stored. When the computer instructions are executed, the above-mentioned method for identifying acupuncture points can be implemented.

[0047] Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0048] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0049] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying acupuncture points, characterized in that: include: Establishing a body surface anatomical landmark coordinate system, and determining the body surface coordinates of human acupuncture points in the body surface anatomical landmark coordinate system; Determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system; Determine the body surface coordinates of the acupuncture points near the acupuncture site coordinates; Calculating the distance between the coordinates of the acupuncture site and the coordinates of the human body acupuncture points near the acupuncture site; The acupoints at the acupuncture site are determined according to the distance and the preset acupoint weights.

2. A method for identifying acupuncture points according to claim 1, characterized in that: The step of establishing a body surface anatomical landmark coordinate system comprises: Acquire a human body surface image by using an image acquisition device; Preprocessing the human body surface image; Inputting the preprocessed human body surface image into the trained first OpenPose model to obtain the coordinates of key points of the human body; A body surface anatomical landmark coordinate system is established according to the coordinates of the key points of the human body and the predetermined body surface anatomical landmark positions and the human body coordinate system.

3. A method for identifying acupuncture points according to claim 2, characterized in that: The preprocessing of the human body surface image comprises: Converting the color space of the human body surface image to HSV; The human body contour in the human body surface image is determined by the YOLOv7 model and the Mask R-CNN model.

4. The method for identifying acupuncture points according to claim 1, characterized in that: The step of determining the body surface coordinates of the human acupuncture points in the body surface anatomical landmark coordinate system comprises: Obtaining the surface anatomical landmarks corresponding to the target acupuncture points; The body surface coordinates of the human acupuncture points are determined based on the bone measurement method and through the body surface anatomical landmarks corresponding to the target acupuncture points.

5. The method for identifying acupuncture points according to claim 1, characterized in that: The step of determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system comprises: Acquire acupuncture action video through image acquisition equipment; Inputting the acupuncture action video into a trained acupuncture gesture recognition model to obtain acupuncture action occurrence frames; According to the acupuncture action occurrence frame, the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system are determined.

6. A method for identifying acupuncture points according to claim 5, characterized in that: Training acupuncture gesture recognition model, including: Get the acupuncture action training video; Preprocessing the acupuncture action training video; Extracting images when acupuncture action occurs and images when no acupuncture action occurs from the acupuncture action training video, and manually labeling them to obtain the category of each acupuncture action; Using the trained second OpenPose model to extract the image when the acupuncture action occurs, and obtain the coordinates of the key points of the hand; Calculating the characteristic vector of the acupuncture action according to the coordinates of the key points of the hand; According to the coordinates of the key points of the hand and the feature vectors of the acupuncture action, the data annotates the characteristics of each acupuncture action; Based on the images when the acupuncture action occurs and the images when no acupuncture action occurs, as well as the category and features of each acupuncture action, a labeled acupuncture action dataset is constructed; Based on the acupuncture action data set, an acupuncture gesture recognition model is trained to obtain a trained acupuncture gesture recognition model.

7. The method for identifying acupuncture points according to claim 5, characterized in that: The step of determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system according to the acupuncture action occurrence frame comprises: Determine the three-dimensional coordinates of the physician's fingertip in the acupuncture action occurrence frame in the body surface anatomical landmark coordinate system; The three-dimensional coordinates of the doctor's fingertip in the body surface anatomical landmark coordinate system are used as the acupuncture site coordinates.

8. A system for identifying acupuncture points, characterized in that: include: A body surface coordinate determination module for acupuncture points on the human body, used to establish a body surface anatomical landmark coordinate system and determine the body surface coordinates of acupuncture points on the human body in the body surface anatomical landmark coordinate system; A module for determining the coordinates of the acupuncture site, used for determining the coordinates of the acupuncture site in the body surface anatomical landmark coordinate system; A neighboring acupoint determination module is used to determine the body surface coordinates of human acupoints near the coordinates of the acupuncture site; A distance calculation module, used to calculate the distance between the coordinates of the acupuncture site and the coordinates of the human body acupoints near the acupuncture site; The acupuncture site acupoint determination module is used to determine the acupoints at the acupuncture site according to the distance and the preset acupoint weights.

9. A device comprising a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, characterized in that: When the processor runs the computer instructions, the acupuncture point identification method according to any one of claims 1 to 7 is executed.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed, the acupuncture point identification method according to any one of claims 1 to 7 can be implemented.

Images