Intelligent acupuncture and moxibustion equipment based on rosemary-rosemary reflection and positioning method

Abstract

The invention discloses intelligent acupuncture and moxibustion equipment based on rosemary-rosemary reflection and a positioning method, relates to the field of intelligent positioning, and is used for solving the problem that the accuracy is reduced due to the fact that a traditional Chinese medicine bone degree division method and a user body type ratio are in a fuzzy relation and positioning of remaining acupuncture points cannot be accurately determined. Preprocessing the images of each partial region of the human body, synthesizing the image definition of each partial region image, the positioning distance difference between the virtual body position and the proportion size and a support vector machine model constructed by a virtual body position acupuncture point position model initial training set, and adjusting the model difference; weighted calculation is carried out through the distance difference between the electrode plate and the preset acupoint position and the preset current difference of the electrode plate, the weighted calculation is compared with the adjusting threshold value, the acupoint positioning accuracy is improved, the application scene of the intelligent acupuncture and moxibustion equipment is improved, and the personalized requirements of different users are met.

Description

Technical Field

[0001] The present invention relates to the field of intelligent positioning, and more specifically, to an intelligent acupuncture device and a positioning method based on vagal-vagal reflex. Background Art

[0002] With the rapid development of medical technology and the increasing awareness of health among people, traditional acupuncture therapy has gradually been valued in modern medicine, and has shown remarkable effects in relieving pain and promoting recovery. In particular, the vagus nerve, as the largest autonomic nerve in the human body, is involved in regulating multiple physiological functions such as cardiovascular, immune and digestive functions. At present, vagus nerve electrical stimulation technology has made certain research progress, and intelligent acupuncture is used to stimulate specific acupoints to meet user needs.

[0003] The prior art has the following deficiencies:

[0004] At present, by integrating advanced image recognition, deep learning and virtual human model alignment technologies, acupoint positioning is accurately achieved, thereby improving the accuracy and effect of treatment and ensuring the accuracy of acupuncture treatment. However, after confirming that an acupoint is correctly positioned, the traditional Chinese medicine bone measurement method is in a fuzzy relationship with the user's body size ratio, and the remaining acupoints cannot be accurately determined, resulting in reduced accuracy. Therefore, an intelligent acupuncture device and positioning method based on the vagus-vagus reflex are proposed.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an intelligent acupuncture device and a positioning method based on the vagal-vagal reflex, which solve the problems raised in the above-mentioned background technology by using different product inspection methods.

[0007] To achieve the above object, the present invention provides the following technical solution: an intelligent positioning method based on vagus-vagus reflex, comprising the following steps:

[0008] S1: collecting images of various parts of the human body and preprocessing the images of various parts of the human body;

[0009] S2: Performing image clarity analysis on the preprocessed images of each part of the human body and performing differential analysis on the virtual body position acupoints and proportional size acupoints to obtain the image clarity of each part of the human body and the virtual body position and proportional size positioning distance difference;

[0010] S3: Based on the initial training set of the virtual body acupoint model, image clarity, and the virtual body position and proportional size positioning distance difference, a support vector machine model is constructed to identify the human body acupoint location;

[0011] S4: Conduct a comparison experiment between the output of the support vector machine model and the actual acupuncture points of the human body to determine the analysis differences and select the corresponding mode to improve the model.

[0012] In a preferred embodiment, in step S1, a real-time image is obtained in an image display device according to the prompts of the user, and then the image is projected to a new viewing plane, and the corresponding positions of the four image vertices are changed so that the acupuncture point area reaches a horizontal viewing angle, and then an intelligent cropping algorithm is used to automatically identify the position of the human body in the image and crop this area out. During cropping, the human body's arms, shoulders, waist, back, thighs, and calves are cropped to obtain images of various parts of the human body.

[0013] In a preferred embodiment, in step S2, after preprocessing the images of various partial regions of the human body, the images of various partial regions are analyzed, and differentiation analysis and image clarity analysis are performed on the virtual body position acupoints and the proportional size acupoints.

[0014] In a preferred embodiment, the image clarity is calculated by weighted analysis based on the edge intensity standard deviation, the image variance after Laplace transformation, the image contrast and the blur value;

[0015] Differential analysis of virtual body acupoints and proportional size acupoints refers to the comparative differences between acupoints obtained through different positioning logics;

[0016] The virtual body position acupuncture points and the proportional size acupuncture points are respectively substituted into the user's three-dimensional coordinate system. The coordinate difference of the two points is calculated through the three-dimensional coordinate system to obtain the virtual body position and proportional size positioning distance difference.

[0017] In a preferred embodiment, in step S3, a support vector machine model is constructed based on the image clarity of each partial area image and the virtual body position and proportional size positioning distance difference, the image clarity of each partial area image is merged into a clear data set, the virtual body position and proportional size positioning distance difference of each partial area image is merged into a difference data set, and each acupoint in the initial training set of the virtual body position acupoint model is collected and merged into a control data set.

[0018] In a preferred embodiment, in step S3, the specific steps of constructing a support vector machine model are as follows:

[0019] Step A1: inputting the clear data set and the difference data set as analysis features, and inputting the control data set as comparison features;

[0020] Step A2: Select a kernel function and calculate the kernel function result according to the analysis characteristics;

[0021] Step A3: setting the adjustment coefficient to converge the kernel function result and calculating the accurate discrimination threshold of acupoint location;

[0022] Step A4: Determine the accuracy of human acupuncture point positioning;

[0023] Step A5: Output the judgment result.

[0024] In a preferred embodiment, in step S4, an evaluation module is provided in the support vector machine model to perform accuracy evaluation on the output results of the support vector machine model to determine the analysis difference.

[0025] In a preferred embodiment, step B1: constructing a data set, arranging the images according to the sequence numbers of the partial regions and merging the discrimination results into a discrimination data set;

[0026] Step B2: Setting evaluation rules;

[0027] Step B3: Perform cross validation and calculate the predicted value and the true value;

[0028] Step B4: Evaluate the accuracy, calculate the ratio of the predicted value to the true value, and use the ratio as the accuracy;

[0029] Step B5: Determine the analysis difference, set the expected interval, and compare whether the accuracy is within the set expected interval to determine its accuracy;

[0030] Step B6: Select a mode to improve the model and use different modes to improve the model according to the accuracy.

[0031] In a preferred embodiment, after the user puts on the wearable elastic cloth belt, the separated electrode sheets built into the cloth bag, each electrode sheet corresponds to a preset acupuncture point, and after the elastic cloth belt is bound, the distance difference between the electrode sheet and the preset acupuncture point and the preset current difference of the electrode sheet are collected;

[0032] The distance difference between the electrode sheet and the preset acupuncture point and the preset current difference of the electrode sheet are obtained by respectively determining the position of the electrode sheet after binding and the determined acupuncture point;

[0033] The distance difference between the electrode and the preset acupuncture point and the preset current difference of the electrode are weighted and calculated, and compared with the adjustment threshold. If it is greater than or equal to the adjustment threshold, the electrode currently close to the corresponding acupuncture point is derived for electrical stimulation treatment. Otherwise, no adjustment is made.

[0034] An intelligent acupuncture device based on vagal-vagal reflex includes a head display device, a wearable elastic cloth belt and a handheld electrical stimulation box;

[0035] The head-mounted display device is used to collect images of various parts of the human body and pre-process the images of various parts of the human body;

[0036] The head display device includes a camera, a display, and a user prompt box;

[0037] Among them, the camera is used to obtain original image data;

[0038] The display is used to display the acquired original image data and display the acupoints in a mixed reality mode or the virtual positioning of the acupoints for autonomous acupoint selection;

[0039] The user prompt box is used to prompt the user to adjust to a suitable posture, such as the abdominal image acquisition contour line prompt box and the thigh image acquisition contour line prompt box;

[0040] The handheld electrical stimulation box is used to integrate the image clarity of each partial area image and the virtual body position and proportional size positioning distance difference with the support vector machine model constructed by the initial training set of the virtual body position acupoint location model and adjust the model difference to determine the acupoint location;

[0041] The handheld electrical stimulation box consists of the following parts:

[0042] As a control and power supply unit, the built-in battery and the external charging port send out an electrical stimulation signal control circuit; an LED screen (11) is arranged on the handheld electrical stimulation box (10) and is used to display treatment parameters; a current adjustment knob (12) is arranged on the handheld electrical stimulation box (10) and is used to adjust the intensity of the electrical stimulation; a switch (13) is used to control the start and stop of the electrical stimulation; a time adjustment knob (14) is arranged on the handheld electrical stimulation box (10) and is used to set the treatment time of the electrical stimulation; a charging port (15) is used to charge the battery built in the handheld electrical stimulation box (10); a main switch (16) is used to control the power on and off of the entire device; a waveform selection button (17) is arranged on the top and is used to select different electrical stimulation waveforms; an electrode sheet buckle (18) is used to connect the fixed separated electrode sheet (6) and the handheld electrical stimulation box (10) to ensure that the electrical signal can be transmitted;

[0043] The wearable elastic cloth belt is used to be worn to the corresponding position of the user for electrical stimulation;

[0044] The wearable elastic band consists of the following parts:

[0045] The metal ring (1) is used to cooperate with the strap to fix the device on the wearer's limb; the blended elastic fabric (2) is a one-piece square design, and a separated electrode sheet (6) is arranged on the inner side, which can cover most of the acupoint area under the fabric, and is used to select the electrode sheet on or near the acupoint for electrical stimulation according to the treatment needs; the separated electrode sheet (6) is arranged on the inner side of the blended elastic fabric (2), and can selectively stimulate specific acupoints according to the treatment needs; the self-adhesive Velcro (thorny hair) (4) and the self-adhesive Velcro (round hair) (5) are used to fix the blended elastic fabric (2) on the wearer's limb; the strap (8) is arranged on the blended elastic fabric (2). One end of the cloth belt (2) is used to pass around the metal ring (2) and cooperate with the Velcro to fix the device; the external electrode (7) is connected to the handheld electrical stimulation box through a double-ended detachable wire (9) to expand the treatment area; the double-ended detachable wire (9) is used to connect the external electrode (7) and the separated electrode sheet (6) to achieve the transmission of electrical stimulation signals; the device also includes a detachable spare sheet, which is also provided with a separated electrode sheet and wires. When it is necessary to increase the treatment area, the spare sheet can be connected to the main body through the metal ring (2), and the corresponding wires can be connected to the external electrode (7) to adapt to the use of people with different limb circumferences.

[0046] Technical effects and advantages of the present invention:

[0047] 1. The present invention collects images of various parts and regions of the human body for preprocessing, performs image clarity analysis on the preprocessed images of various parts and regions of the human body, and performs differential analysis on the virtual body position acupoints and proportional size acupoints of each partial region image, constructs and adjusts a support vector machine through the image clarity of each partial region image and the virtual body position and proportional size positioning distance difference, performs human body acupoint positioning discrimination according to the support vector machine model, performs a comparison experiment on the discrimination results and the actual acupoints of the human body, determines the analysis difference, selects the corresponding mode according to the analysis difference to improve the model, further improves the recognition of human body acupoints, solves the fuzzy relationship between the traditional Chinese medicine bone measurement method and the user's body size ratio, and improves the accuracy of acupoint positioning.

[0048] 2. The present invention performs weighted calculation through the distance difference between the electrode sheet and the preset acupuncture points and the preset current difference of the electrode sheet, and compares it with the adjustment threshold, thereby improving the applicable scenarios of the intelligent acupuncture device and meeting the personalized needs of different users. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 The present invention is a schematic diagram of an intelligent positioning method based on the vagus-vagus reflex.

[0050] Figure 2 The present invention is a method flow chart of an intelligent positioning method based on the vagus-vagus reflex.

[0051] Figure 3 These are the outer and inner side views of a wearable elastic cloth belt of an intelligent acupuncture device based on the vagal-vagal reflex of the present invention.

[0052] Figure 4 This is an example diagram of on-demand lengthening of a wearable elastic band of an intelligent acupuncture device based on the vagal-vagal reflex of the present invention.

[0053] Figure 5 This is an example diagram of a handheld electrical stimulation box of an intelligent acupuncture device based on the vagal-vagal reflex of the present invention.

[0054] Figure 6 This is an example diagram of an application scenario of an intelligent acupuncture device based on the vagal-vagal reflex of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0056] Example 1

[0057] See also Figure 1 to Figure 2 , an intelligent positioning method based on vagus-vagus reflex, the specific operation process is as follows:

[0058] S1: collecting images of various parts of the human body and preprocessing the images of various parts of the human body;

[0059] S2: Performing image clarity analysis on the preprocessed images of each part of the human body and performing differential analysis on the virtual body position acupoints and proportional size acupoints to obtain the image clarity of each part of the human body and the virtual body position and proportional size positioning distance difference;

[0060] S3: Based on the initial training set of the virtual body acupoint model, image clarity, and the virtual body position and proportional size positioning distance difference, a support vector machine model is constructed to identify the human body acupoints;

[0061] S4: Perform a comparison experiment between the output of the support vector machine model and the actual acupuncture points of the human body to determine the analysis differences and select the corresponding mode to improve the model;

[0062] The specific implementation is as follows:

[0063] In step S1, a real-time image obtained by the user according to the prompt is obtained in the image display device, each frame of the image is processed, the noise in the image is removed by using a Gaussian filter, and then the image is projected to a new viewing plane, the corresponding positions of the four image vertices are changed, so that the acupuncture point area reaches a horizontal viewing angle, and then an intelligent cropping algorithm is used to automatically identify the position of the human body in the image, and this part of the area is cropped out. Furthermore, when cropping, the human body's arms, shoulders, waist, back, thighs, and calves are cropped to obtain images of various parts of the human body;

[0064] It should be noted that the images of various parts of the human body can be obtained by transmitting the image data of various parts of the human body in the recording device at the corresponding stage into the system. When the images of various parts of the human body are recorded using the recording device, each area is recorded in turn for the subsequent division of the images of various parts of the human body.

[0065] Specifically, the intelligent cropping algorithm uses a pre-trained deep learning model; when cropping, the image is first input into the model, and the model will output a heat map of the human body position; then, the center position of the human body is determined according to the maximum value point in the heat map; finally, a region of a predefined size is cropped out with this center position as the center, and then the region of the predefined size is divided, and the images of various parts of the human body are divided according to the characteristic images of the arms, shoulders, waist, back, thighs and calves, which will not be elaborated here;

[0066] Specifically, the preprocessing of the images of various parts of the human body includes uniformly scaling the size of the images to a predefined size, normalizing the pixel values, and performing enhancement operations on the normalized images to improve the generalization ability of the model;

[0067] Pixel value normalization is mainly to uniformly scale the pixel values ​​in the image to a fixed range, reduce the impact of different brightness and contrast of the image, improve the processing stability of the model, and adjust the original image to a predefined fixed size;

[0068] Specifically, assuming that the pixel value range of the image is [0,255], the normalization method is normalized by the [0,1] interval, and the specific formula is expressed as:

[0069]

[0070] In the formula, x is the value of each pixel in the image, x norm is the normalized pixel value, x min is the minimum value of the original pixel value, x max is the maximum value of the original pixel value;

[0071] Perform data augmentation on the normalized images, perform random rotation, flipping, brightness adjustment, etc., to make the model more stable in different lighting, angles, viewing distances, and other environments, thereby enhancing generalization capabilities;

[0072] In step S2, after preprocessing the images of various partial regions of the human body, the images of various partial regions are analyzed, and the virtual body position acupoints and the proportional size acupoints are respectively subjected to differential analysis and image clarity analysis;

[0073] Specifically, image clarity analysis usually extracts clarity features through a series of image processing techniques to quantify the clarity level so as to optimize the location and regional division of acupuncture points. The specific image clarity extraction process is as follows:

[0074] Convert the image to a grayscale image to simplify the calculation. Specifically, a color image usually contains three channels, RGB. The specific formula for converting a color image to a grayscale image is as follows:

[0075] I g =0.2989×I R +0.5870×I G +0.1140×I B

[0076] In the formula, I R ,I G ,I B They represent the red, green, and blue channels of the image respectively. 0.2989, 0.2870, and 0.1140 are the perceptual weights of the three color channels for the human eye. g is a grayscale image;

[0077] It should be noted that converting a color image into a grayscale image can preserve the brightness information and reduce the number of image channels, making the subsequent edge detection algorithm;

[0078] Then use the edge detection algorithm to calculate the edge strength. Taking the Sobel operator as an example, the gradients of the image in the horizontal and vertical directions are calculated respectively. Specifically, two 3×3 convolution kernels are used to calculate the horizontal direction G x and vertical direction G y The gradient is:

[0079] as well as

[0080] Convolve the Sobel convolution kernel with the grayscale image I g The convolution operation is performed to obtain the horizontal gradient map and the vertical gradient map. The specific formula is:

[0081] I x =I g *Gx and I y =I g *G y ;

[0082] In the formula, I x is the horizontal gradient map, I y is the vertical gradient map;

[0083] Calculate the edge strength map. The specific formula is:

[0084]

[0085] Where, the edge strength I edge Each pixel value represents the edge strength at that location. The greater the strength, the more obvious the edge of the area, indicating a higher image clarity.

[0086] The edge intensity standard deviation is obtained for the edge intensity values ​​of all pixels in the edge intensity;

[0087] Apply the Laplace operator to the grayscale image and convolve the grayscale image with the discrete Laplace kernel to obtain the Laplace transformed image:

[0088] I lap =I g *L

[0089] In the formula, I lap is the image after Laplace transformation, I g is a grayscale image, L is a discrete Laplace kernel;

[0090] Count the image after Laplace transformation and substitute it into the variance calculation to obtain the variance of the image after Laplace transformation;

[0091] It should be noted that the larger the image variance, the more high-frequency components there are in the image, that is, the richer the edge and detail information, the clearer the image, which will not be elaborated here;

[0092] Then count the number of pixels at each gray level to get the gray distribution, and then calculate the standard deviation of the gray value to get the image contrast;

[0093] Finally, the proportion of blurred areas in the image is detected, the frequency domain information of the image is extracted using Fourier transform, and the distribution ratio of high-frequency components is analyzed to obtain the blur value. Specifically, the lower the blur, the clearer the image.

[0094] It should be noted that the above-mentioned grayscale conversion, edge algorithm, Laplace transform and Fourier transform are all existing technologies. It can be understood by the experimenters that the above-mentioned existing technologies can be used to analyze the clarity of the image from multiple aspects;

[0095] Furthermore, a comprehensive clarity score, i.e., image clarity, is calculated by weighted analysis based on the edge intensity standard deviation, image variance after Laplace transformation, image contrast, and blur value;

[0096] Among them, the differential analysis of virtual body acupoints and proportional size acupoints refers to the comparative differences between acupoints obtained through different positioning logics;

[0097] Among them, the logic of obtaining virtual body acupuncture points is to determine the acupuncture point locations through a pre-built three-dimensional virtual human model, which scales and adjusts the virtual human body in combination with the user's specific body parameters (such as height, weight, and posture) to make it closer to the user's actual body shape, and uses the standard acupuncture point location coordinates preset in the model to map these coordinates to the user's actual body area, thereby obtaining the virtual body acupuncture points;

[0098] Among them, the logic of obtaining proportional size acupoints is based on the bone measurement method of traditional Chinese medicine, which measures the size and proportion of specific parts of an individual's body and determines the location of acupoints according to the distance standard of Chinese medicine acupoints (such as "inch");

[0099] It should be noted that the bone measurement method of traditional Chinese medicine gives priority to determining an accurate acupuncture point. For example, the Shenque acupuncture point is located in the navel part of the human body. This part can be determined by the collected images and virtual body position. After the Shenque acupuncture point is determined, the remaining acupuncture points are determined according to the bone measurement method of traditional Chinese medicine.

[0100] It is understandable that in the virtual body acupoint model, standard acupoint location coordinates are preset, and these coordinates can be set based on typical human anatomy or acupoint locations, which will not be described in detail here;

[0101] Substitute the virtual body position acupuncture points and the proportional size acupuncture points into the user's three-dimensional coordinate system respectively, calculate the coordinate difference of the two points through the three-dimensional coordinate system, and obtain the virtual body position and proportional size positioning distance difference;

[0102] In step S3, each acupoint in the initial training set of the virtual body position acupoint model is collected and merged into a control data set, and the image clarity of each partial area image is calculated according to the method in step S2 and merged into a clear data set, and the virtual body position and proportional size positioning distance difference of each partial area image is merged into a difference data set;

[0103] Among them, each acupoint in the initial training set in the virtual body position acupoint model is used as a control data set. Specifically, there are significant differences between the initial training set in the virtual body position acupoint model and the existing preset virtual body position acupoints. The specific logic is that if the initial training set in the virtual body position acupoint model can be used satisfactorily, there is no need to continue to optimize to the existing preset virtual body position acupoints;

[0104] The system uses the clear data set, the difference data set and the control data set as input variables to build a support vector machine model. The support vector machine performs simulation tests through the input variables and calculates the accurate discrimination threshold of acupoint positioning. After obtaining the accurate discrimination threshold of acupoint positioning, the system compares the input data of the human body area to be judged with the calculated accurate discrimination threshold of acupoint positioning, and judges the human body acupoint positioning and outputs the judgment result as the output result. The specific steps are as follows:

[0105] Step A1: inputting the clear data set and the difference data set as analysis features, and inputting the control data set as comparison features;

[0106] Step A2: Select the Sigmoid function as the kernel function to transform the input features. The Sigmoid kernel function formula is specifically expressed as:

[0107] K(x,y)=tanh(γx*y+c)

[0108] Where K(x,y) is the kernel function result after feature conversion, tanh is the hyperbolic tangent function, γ is a hyperparameter used to control the width of the kernel function, c is a constant used to control the offset of the kernel function, x and y are the two input features in this example, that is, the corresponding data of the same part of the image in the clear data set and the difference data set;

[0109] Step A3: Set the adjustment coefficient k, and converge the kernel function result according to the adjustment coefficient. The convergence formula can be:

[0110] K new (x,y)=k*K old (x,y)

[0111] In the formula, k is the adjustment coefficient, K new (x, y) is the kernel function result after convergence, K old (x, y) can be set to the sum of the kernel function results obtained after feature conversion of all partial area images, and the converged kernel function result is used as the accurate discrimination threshold for acupoint location;

[0112] Step A4: judging the accuracy of human acupoint positioning, calculating the kernel function result after feature conversion according to the corresponding data of each partial region image in the analysis feature, comparing the calculated kernel function result with the acupoint positioning accuracy discrimination threshold, if the calculated kernel function result is higher than the acupoint positioning accuracy discrimination threshold, then marking the corresponding partial region image as 1; otherwise, marking it as 0;

[0113] Step A5: Output the discrimination result. When the partial area image is marked as 1, the discrimination result is accurate; when the partial area image is marked as 0, the discrimination result is inaccurate, and the discrimination result is output;

[0114] The support vector machine is used to comprehensively analyze the clear data set and the difference data set to judge the accuracy of the images in some areas, which greatly improves the judgment of the degree positioning accuracy. The method of setting the adjustment coefficient facilitates the subsequent improvement and adjustment of the support vector machine model.

[0115] It should be noted that the role of the Sigmoid function is to convert the input features into data that is easy to analyze and process. The selection of the kernel function and the loss function in the above steps is not unique. The hyperparameter γ and the constant c involved in the steps can be set according to the actual situation such as Huang Jing's discrimination requirements. For example, the hyperparameter is set to 0.8, the constant is set to 0.2, and the adjustment coefficient k is preset to 1, etc. The preset adjustment coefficient is analyzed and adjusted by step S4, which will not be repeated here;

[0116] In step S4, an evaluation module is set in the support vector machine model to evaluate the accuracy of the output results of the support vector machine model, determine the analysis difference, and select different modes to improve the model according to the accuracy. The specific steps are as follows:

[0117] Step B1: construct a data set, arrange the discrimination results output by the support vector machine according to the data in the control data set in the comparison feature, that is, arrange them according to the sequence number of the partial area images and merge the discrimination results into a discrimination data set;

[0118] Step B2: Setting evaluation rules, comparing the discrimination data set and the control data set according to the sequence number of the partial area images, if the values ​​are the same and are 1, the partial area image mark is updated to true low positioning accuracy; if the values ​​are the same and are 0, the partial area image mark is updated to true high positioning accuracy; if the values ​​are different and the corresponding data in the control data set is 1, the partial area image mark is updated to false high positioning accuracy; if the values ​​are different and the corresponding data in the control data set is 0, the partial area image mark is updated to false low positioning accuracy;

[0119] Step B3: Cross-validation, select k-fold cross-validation, randomly divide the discriminant data set and the control data set into k subsets of equal size, and correspond one to one according to the sequence number of the partial area images. Select one subset as the test set each time, and the remaining k-1 subsets as the training set. Sum the marks with low true positioning accuracy and high true positioning accuracy in the test set, and divide the sum result by the total number of marks in the test set to obtain the test ratio. The same is true for the training set to obtain the training ratio. Repeat the cross-validation, and calculate the average of the obtained test ratio and training ratio to obtain the predicted value and true value;

[0120] Step B4: Evaluate the accuracy, calculate the ratio of the predicted value to the true value, and use the ratio as the accuracy. It should be explained that the closer the predicted value is to the true value, the more accurate the prediction result is, that is, the smaller the ratio of the predicted value to the true value is than 1, the more accurate the prediction result is;

[0121] Step B5: Determine the analysis difference, set the expected interval, and compare whether the accuracy is within the expected interval. When the accuracy is within the expected interval, the model accuracy is judged to be high; when the accuracy is outside the expected interval, the model accuracy is judged to be low;

[0122] Step B6: Select the mode to improve the model. If the model is judged to be highly accurate, save the construction mechanism of the support vector machine model, that is, save the construction process of the support vector machine model; if the model is judged to be low in accuracy, set the correction parameters. Correct the adjustment coefficient k according to the correction parameters:

[0123]

[0124] Among them, K old is the uncorrected adjustment factor, K new The uncorrected adjustment coefficient is obtained by randomly generating the correction parameter The adjustment coefficient k is corrected, and the support vector machine model operation process is repeated until the accuracy calculated in the evaluation module is within the expected range. The correction is completed and the corrected adjustment coefficient is retained;

[0125] It should be noted that the expected interval in the evaluation module can be set according to the expected results. For example, the expected interval is set to [0.7, 1.5], and the correction parameter The random generation of can also be delimited, for example, the correction parameter Random generation is performed in the interval [0.5, 1.8] to reduce calculation time and improve model operation efficiency.

[0126] The output results of the support vector machine model are evaluated for accuracy through the evaluation module, which secondary improves the accuracy of the model in identifying the human acupoints. The support vector machine model greatly shortens the time for identifying acupoints, and solves the problem of being unable to accurately determine the remaining acupoints, resulting in reduced accuracy.

[0127] Example 2

[0128] In the first embodiment of the present invention, an example is given to illustrate the operation strategy of comprehensively analyzing the image clarity of each partial area image, the virtual body position and the proportional size positioning distance difference, and the support vector machine model constructed by the initial training set of the virtual body position acupoint model, and adjusting the model difference; however, in the first embodiment, only the acupoints are positioned, and the difference between the electrode sheet and the acupoints in the actual application of the intelligent acupuncture device is not analyzed, resulting in unstable acupoint locking and reduced user experience; in view of the above problems, the second embodiment of the present invention is further refined;

[0129] See also Figure 3-Figure 5 , an intelligent acupuncture device based on vagal-vagal reflex, including a head-mounted display device, a wearable elastic cloth belt, and a handheld electrical stimulation box;

[0130] The head-mounted display device is used to collect images of various parts of the human body and pre-process the images of various parts of the human body;

[0131] The handheld electrical stimulation box is used to integrate the image clarity of each partial area image and the virtual body position and proportional size positioning distance difference with the support vector machine model constructed by the initial training set of the virtual body position acupoint location model and adjust the model difference to determine the acupoint location;

[0132] The wearable elastic cloth belt is used to be worn to the corresponding position of the user for electrical stimulation;

[0133] After the user puts on the wearable elastic cloth belt, the separated electrode sheets built into the cloth bag correspond to the preset acupuncture points. After the elastic cloth belt is bound, the distance difference between the electrode sheet and the preset acupuncture points and the preset current difference of the electrode sheet are collected;

[0134] Specifically, the distance difference between the electrode sheet and the preset acupuncture point and the preset current difference of the electrode sheet are obtained by respectively determining the position of the electrode sheet after binding and the determined acupuncture point, which will not be described in detail here;

[0135] The distance difference between the electrode sheet and the preset acupuncture point and the preset current difference of the electrode sheet are weighted and calculated, and compared with the adjustment threshold. If it is greater than or equal to the adjustment threshold, the electrode sheet currently close to the corresponding acupuncture point is derived for electrical stimulation treatment. Otherwise, no adjustment is made, and the words "the current wearing position is not good" are generated and sent to the handheld electrical stimulation box;

[0136] Example 3

[0137] See also Figures 3 to 6 , an intelligent acupuncture device based on vagal-vagal reflex, including a head-mounted display device, a wearable elastic cloth belt and a handheld electrical stimulation box;

[0138] The head display device includes a camera, a display, and a user prompt box;

[0139] Among them, the camera is used to obtain original image data;

[0140] The display is used to display the acquired original image data and display the acupoints in a mixed reality mode or the virtual positioning of the acupoints for autonomous acupoint selection;

[0141] The user prompt box is used to prompt the user to adjust to a suitable posture, such as the abdominal image acquisition contour line prompt box and the thigh image acquisition contour line prompt box.

[0142] The wearable elastic belt consists of the following parts:

[0143] The metal ring (1) is used to cooperate with the strap to fix the device on the wearer's limb; the blended elastic fabric (2) is a one-piece square design, and a separated electrode sheet (6) is arranged on the inner side, which can cover most of the acupoint area under the fabric, and is used to select the electrode sheet on or near the acupoint for electrical stimulation according to the treatment needs; the separated electrode sheet (6) is arranged on the inner side of the blended elastic fabric (2), and can selectively stimulate specific acupoints according to the treatment needs; the self-adhesive Velcro (thorny hair) (4) and the self-adhesive Velcro (round hair) (5) are used to fix the blended elastic fabric (2) on the wearer's limb; the strap (8) is arranged on the blended elastic fabric (2). One end of the cloth belt (2) is used to pass around the metal ring (2) and cooperate with the Velcro to fix the device; the external electrode (7) is connected to the handheld electrical stimulation box through a double-ended detachable wire (9) to expand the treatment area; the double-ended detachable wire (9) is used to connect the external electrode (7) and the separated electrode sheet (6) to achieve the transmission of electrical stimulation signals; the device also includes a detachable spare sheet, which is also provided with a separated electrode sheet and wires. When it is necessary to increase the treatment area, the spare sheet can be connected to the main body through the metal ring (2), and the corresponding wires can be connected to the external electrode (7) to adapt to the use of people with different limb circumferences.

[0144] The handheld electrical stimulation box consists of the following parts:

[0145] As a control and power supply unit, the device has a built-in battery and an external charging port for sending out an electrical stimulation signal control circuit; an LED screen (11) is arranged on the handheld electrical stimulation box (10) for displaying treatment parameters; a current adjustment knob (12) is arranged on the handheld electrical stimulation box (10) for adjusting the intensity of the electrical stimulation; a switch (13) is used to control the start and stop of the electrical stimulation; a time adjustment knob (14) is arranged on the handheld electrical stimulation box (10) for setting the treatment time of the electrical stimulation; a charging port (15) is used to charge the battery built in the handheld electrical stimulation box (10); a main switch (16) is used to control the power on and off of the entire device; a waveform selection button (17) is arranged on the top for selecting different electrical stimulation waveforms; and an electrode sheet buckle (18) is used to connect the fixed separated electrode sheet (6) and the handheld electrical stimulation box (10) to ensure that the electrical signal can be transmitted.

[0146] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0147] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0148] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0149] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0150] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0151] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0152] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0153] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0154] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0155] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent positioning method based on vagal-vagal reflex, characterized in that: The following steps are included: S1: collecting images of various parts of the human body and preprocessing the images of various parts of the human body; S2: Performing image clarity analysis on the preprocessed images of each part of the human body and performing differential analysis on the virtual body position acupoints and proportional size acupoints to obtain the image clarity of each part of the human body and the virtual body position and proportional size positioning distance difference; S3: Based on the initial training set of the virtual body acupoint model, image clarity, and the virtual body position and proportional size positioning distance difference, a support vector machine model is constructed to identify the human body acupoints; S4: Conduct a comparison experiment between the output of the support vector machine model and the actual acupuncture points of the human body to determine the analysis differences and select the corresponding mode to improve the model.

2. The intelligent positioning method based on vagal-vagal reflex according to claim 1, characterized in that: In step S1, a real-time image is obtained in the image display device according to the prompts of the user, and then the image is projected to a new viewing plane, and the corresponding positions of the four image vertices are changed so that the acupuncture point area reaches a horizontal viewing angle, and then an intelligent cropping algorithm is used to automatically identify the position of the human body in the image and crop this area out. When cropping, the human body is cropped according to the arms, shoulders, waist, back, thighs, and calves to obtain images of various parts of the human body.

3. The intelligent positioning method based on vagal-vagal reflex according to claim 2, characterized in that: In step S2, after preprocessing the images of various partial regions of the human body, the images of various partial regions are analyzed, and differentiation analysis and image clarity analysis are performed on the virtual body position acupoints and the proportional size acupoints.

4. The intelligent positioning method based on vagal-vagal reflex according to claim 3, characterized in that: The image clarity is calculated by weighted analysis based on the edge intensity standard deviation, image variance after Laplace transformation, image contrast, and blur value; Differential analysis of virtual body acupoints and proportional size acupoints refers to the comparative differences between acupoints obtained through different positioning logics; The virtual body position acupuncture points and the proportional size acupuncture points are respectively substituted into the user's three-dimensional coordinate system. The coordinate difference of the two points is calculated through the three-dimensional coordinate system to obtain the virtual body position and proportional size positioning distance difference.

5. The intelligent positioning method based on vagal-vagal reflex according to claim 1, characterized in that: In step S3, a support vector machine model is constructed based on the image clarity of each partial area image and the virtual body position and proportional size positioning distance difference, the image clarity of each partial area image is merged into a clear data set, the virtual body position and proportional size positioning distance difference of each partial area image is merged into a difference data set, and each acupoint in the initial training set of the virtual body position acupoint model is collected and merged into a control data set.

6. The intelligent positioning method based on vagal-vagal reflex according to claim 5, characterized in that: In step S3, the specific steps of building a support vector machine model are as follows: Step A1: inputting the clear data set and the difference data set as analysis features, and inputting the control data set as comparison features; Step A2: Select a kernel function and calculate the kernel function result according to the analysis characteristics; Step A3: setting the adjustment coefficient to converge the kernel function result and calculating the accurate discrimination threshold of acupoint location; Step A4: Determine the accuracy of human acupuncture point positioning; Step A5: Output the judgment result.

7. The intelligent positioning method based on vagal-vagal reflex according to any one of claim 6, characterized in that: In step S4, an evaluation module is set in the support vector machine model to evaluate the accuracy of the output results of the support vector machine model and determine the analysis difference.

8. The intelligent positioning method based on vagal-vagal reflex according to any one of claim 7, characterized in that: Step B1: construct a data set, arrange the images according to the sequence number of the partial regions and merge the discrimination results into a discrimination data set; Step B2: Setting evaluation rules; Step B3: Perform cross validation and calculate the predicted value and the true value; Step B4: Evaluate the accuracy, calculate the ratio of the predicted value to the true value, and use the ratio as the accuracy; Step B5: Determine the analysis difference, set the expected interval, and compare whether the accuracy is within the set expected interval to determine its accuracy; Step B6: Select a mode to improve the model and use different modes to improve the model according to the accuracy.

9. The intelligent positioning method based on vagal-vagal reflex according to any one of claim 1, characterized in that: After the user puts on the wearable elastic cloth belt, the separated electrode sheets built into the cloth bag correspond to the preset acupuncture points. After the elastic cloth belt is bound, the distance difference between the electrode sheet and the preset acupuncture points and the preset current difference of the electrode sheet are collected; The distance difference between the electrode sheet and the preset acupuncture point and the preset current difference of the electrode sheet are obtained by respectively determining the position of the electrode sheet after binding and the determined acupuncture point; The distance difference between the electrode and the preset acupuncture point and the preset current difference of the electrode are weighted and calculated, and compared with the adjustment threshold. If it is greater than or equal to the adjustment threshold, the electrode currently close to the corresponding acupuncture point is derived for electrical stimulation treatment. Otherwise, no adjustment is made.

10. An intelligent acupuncture device based on vagal-vagal reflex, used to implement an intelligent positioning method based on vagal-vagal reflex as claimed in any one of claims 1 to 9, characterized in that: It includes a head-mounted display device, a wearable elastic cloth belt, and a handheld electrical stimulation box; The head-mounted display device is used to collect images of various parts of the human body and pre-process the images of various parts of the human body; The head display device includes a camera, a display, and a user prompt box; Among them, the camera is used to obtain original image data; The display is used to display the acquired original image data and display the acupoints in a mixed reality mode or the virtual positioning of the acupoints for autonomous acupoint selection; The user prompt box is used to prompt the user to adjust to a suitable posture, such as the abdominal image acquisition contour line prompt box and the thigh image acquisition contour line prompt box; The handheld electrical stimulation box is used to integrate the image clarity of each partial area image and the virtual body position and proportional size positioning distance difference with the support vector machine model constructed by the initial training set of the virtual body position acupoint location model and adjust the model difference to determine the acupoint location; The handheld electrical stimulation box consists of the following parts: As a control and power supply unit, the built-in battery and the external charging port send out an electrical stimulation signal control circuit; an LED screen (11) is arranged on the handheld electrical stimulation box (10) and is used to display treatment parameters; a current adjustment knob (12) is arranged on the handheld electrical stimulation box (10) and is used to adjust the intensity of the electrical stimulation; a switch (13) is used to control the start and stop of the electrical stimulation; a time adjustment knob (14) is arranged on the handheld electrical stimulation box (10) and is used to set the treatment time of the electrical stimulation; a charging port (15) is used to charge the battery built in the handheld electrical stimulation box (10); a main switch (16) is used to control the power on and off of the entire device; a waveform selection button (17) is arranged on the top and is used to select different electrical stimulation waveforms; an electrode sheet buckle (18) is used to connect the fixed separated electrode sheet (6) and the handheld electrical stimulation box (10) to ensure that the electrical signal can be transmitted; The wearable elastic cloth belt is used to be worn to the corresponding position of the user for electrical stimulation; The wearable elastic band consists of the following parts: The metal ring (1) is used to cooperate with the strap to fix the device on the wearer's limb; the blended elastic fabric (2) is a one-piece square design, and a separated electrode sheet (6) is arranged on the inner side, which can cover most of the acupoint area under the fabric, and is used to select the electrode sheet on or near the acupoint for electrical stimulation according to the treatment needs; the separated electrode sheet (6) is arranged on the inner side of the blended elastic fabric (2), and can selectively stimulate specific acupoints according to the treatment needs; the self-adhesive Velcro (thorny hair) (4) and the self-adhesive Velcro (round hair) (5) are used to fix the blended elastic fabric (2) on the wearer's limb; the strap (8) is arranged on the blended elastic fabric (2). One end of the cloth belt (2) is used to pass around the metal ring (2) and cooperate with the Velcro to fix the device; the external electrode (7) is connected to the handheld electrical stimulation box through a double-ended detachable wire (9) to expand the treatment area; the double-ended detachable wire (9) is used to connect the external electrode (7) and the separated electrode sheet (6) to achieve the transmission of electrical stimulation signals; the device also includes a detachable spare sheet, which is also provided with a separated electrode sheet and wires. When it is necessary to increase the treatment area, the spare sheet can be connected to the main body through the metal ring (2), and the corresponding wires can be connected to the external electrode (7) to adapt to the use of people with different limb circumferences.

Images