An extremely weak magnetic measurement method and system for integrated traditional Chinese and Western medicine acupuncture departments

Abstract

The present invention relates to the technical field of human signal measurement, and particularly relates to an extremely weak magnetic measurement method and system for a combined traditional Chinese and Western medicine acupuncture department. Based on the characteristic that the angle and depth of needle insertion have a certain influence on the magnetic field change of acupoints, the present application conducts commonality analysis and universality analysis on the classified magnetic field intensity data at each needle insertion angle and each needle insertion depth, so as to determine the corresponding influencing factors, that is, the degree of influence; thereby, based on the influencing factors, the weights of the magnetic field intensity in the iterative classification process are adaptively adjusted, making the update of the weights in the weak classifier training process more in line with objective facts, thereby improving the accuracy and convergence speed of the human acupoint acupuncture model constructed according to each weak classifier, and making the accuracy of extremely weak magnetic measurement higher.

Description

Technical Field

[0001] The present invention relates to the technical field of human signal measurement, and particularly relates to an extremely weak magnetic measurement method and system for a combined traditional Chinese and Western medicine acupuncture department. Background Art

[0002] Acupuncture is a traditional Chinese medicine treatment behavior that uses certain needles to pierce specific parts of the human body for stimulation to achieve treatment. Inexperienced physicians often cannot achieve good treatment effects when performing acupuncture on patients. Currently, the acupuncture process mainly relies on the experience of physicians for treatment, resulting in uneven acupuncture levels. During the acupuncture process, the stimulation of the needles on the human body will generate an electrical response signal, which in turn triggers an extremely weak magnetic field locally. Therefore, the actual situation of acupuncture can be measured based on the human signals reflected by the magnetic field.

[0003] Therefore, usually, the extremely weak magnetic field measured at the acupuncture points during the acupuncture process is compared with the standard data to determine whether the corresponding magnetic field intensity data belongs to abnormal data, thereby indirectly assisting in measuring the abnormal situation during the acupuncture process. For the standard data, the prior art usually uses machine learning methods to train weak classifiers through the obtained magnetic field intensity data, classify and summarize the magnetic field intensity data of different acupuncture points, establish a human acupuncture point model, and then compare the output result of the human acupuncture point model with the measured extremely weak magnetic field to perform more accurate extremely weak magnetic measurement.

[0004] However, the process of the human acupuncture point model in the prior art generally only uses the acupoint difference and the corresponding acupoint magnetic field intensity for establishment, and corrects and adjusts the weight using the classification error of the classifier during the iteration of the weak classifier. This establishment method is likely to result in too high a misclassification rate and a slow convergence speed, making the accuracy and efficiency of the constructed human acupuncture point model poor, resulting in low accuracy and efficiency of extremely weak magnetic measurement. Summary of the Invention

[0005] In order to solve the technical problem that the process of the human acupuncture point model in the prior art generally only uses the acupoint difference and the corresponding acupoint magnetic field intensity for establishment, resulting in poor accuracy and efficiency of the constructed human acupuncture point model and low accuracy of extremely weak magnetic measurement, the purpose of this application is to provide an extremely weak magnetic measurement method and system for a combined traditional Chinese and Western medicine acupuncture department. The specific technical solutions adopted are as follows:

[0006] This application proposes an extremely weak magnetic measurement method for a combined traditional Chinese and Western medicine acupuncture department, and the method includes:

[0007] During the acupuncture of a patient by combined traditional Chinese and Western medicine, collect the magnetic field intensity data under each acupuncture point and the acupuncture angle and depth when each magnetic field intensity data is collected;

[0008] Based on all the magnetic field intensity data under each acupuncture point, construct the corresponding basic weak classifier; during the classification training of the basic weak classifier, determine the correct magnetic field intensity data representing correct classification and the incorrect magnetic field intensity data representing incorrect classification corresponding to each iteration classification process of each acupuncture point; determine the classification error rate of each iteration classification process according to the quantity and weight distribution of the incorrect magnetic field intensity data.

[0009] Determine the data commonality of each acupuncture angle according to the fluctuation and overall quantity of the correct magnetic field intensity data corresponding to each acupuncture angle; determine the data universality of each acupuncture angle according to the overall quantity difference and overall numerical difference between the incorrect magnetic field intensity data and the correct magnetic field intensity data corresponding to each acupuncture angle; determine the influence factor of each acupuncture angle according to the relative magnitude between the data commonality and the data universality; determine the influence factor of each acupuncture depth.

[0010] During each iteration classification process, determine the magnetic field intensity weight of each magnetic field intensity data after each iteration classification process according to the classification error rate, the influence factor of the acupuncture angle and the influence factor of the acupuncture depth corresponding to each magnetic field intensity data; construct the human acupuncture model of each acupuncture point according to the weak classifiers obtained during the iteration process of the magnetic field intensity weight; perform extremely weak magnetic measurement according to the human acupuncture model.

[0011] Furthermore, the acquisition process of the basic weak classifier includes:

[0012] Take the reciprocal of the number of magnetic field intensity data under each acupuncture point as the initial weight of each magnetic field intensity data; based on the initial weight, construct a decision tree for all the magnetic field intensity data under each acupuncture point through the decision tree algorithm; use the decision tree as the basic weak classifier of each acupuncture point.

[0013] Furthermore, the acquisition process of the classification error rate includes:

[0014] Take the weight of each magnetic field intensity data before each iteration classification process as the reference weight; take the ratio between the cumulative value of the reference weights of all the correct magnetic field intensity data corresponding to each acupuncture point in each iteration classification process and the cumulative value of the reference weights of all the magnetic field intensity data as the classification error rate of each acupuncture point in each iteration classification process.

[0015] Furthermore, the acquisition process of the data commonality includes:

[0016] In each iteration classification process of each acupuncture point, the ratio between the number of correct magnetic field intensity data corresponding to each acupuncture angle and the total number of correct magnetic field intensity data is used as the reference correct proportion for each acupuncture angle.

[0017] Normalize the product of the negative correlation mapping value of the variance of the correct magnetic field intensity data corresponding to each acupuncture angle and the reference correct proportion to determine the data commonality corresponding to each acupuncture angle in each iteration classification process of each acupuncture point.

[0018] Furthermore, the process of obtaining the data universality includes:

[0019] At each acupuncture angle, perform a negative correlation mapping on the difference between the mean of all correct magnetic field intensity data and the mean of all incorrect magnetic field intensity data to determine the corresponding intensity universality; perform a negative correlation mapping on the difference between the total number of correct magnetic field intensity data and the total number of incorrect magnetic field intensity data to determine the corresponding quantity universality; normalize the product between the intensity universality and the quantity universality to determine the data universality corresponding to each acupuncture angle in each iteration classification process of each acupuncture point.

[0020] Furthermore, the process of obtaining the influencing factor includes:

[0021] Determine the influencing factor corresponding to each acupuncture angle in each iteration classification process of each acupuncture point according to the product of the data universality and the data commonality.

[0022] Furthermore, the process of obtaining the magnetic field intensity weight includes:

[0023] In each iteration classification process of each acupuncture point, the product of the normalized value of the classification error rate and the indication coefficient corresponding to each magnetic field intensity data is used as the initial weight update parameter for each magnetic field intensity data; among them, the indication coefficient of the correct magnetic field intensity data is 1, and the indication coefficient of the incorrect magnetic field intensity data is 0.

[0024] The sum value between the influencing factor of the acupuncture angle corresponding to each magnetic field intensity data and the influencing factor of the corresponding acupuncture depth is used as the reference sum value; the product of the negative correlation mapping value of the reference sum value and the initial weight update parameter is used as the corrected weight update parameter.

[0025] The product of the negative correlation mapping value of the corrected weight update parameter and the reference weight is used as the magnetic field intensity weight in each iteration classification process of each acupuncture point.

[0026] Furthermore, the process of obtaining the human acupuncture point acupuncture model includes:

[0027] Under each acupuncture point, an iterative weak classifier corresponding to each iterative classification process is determined according to the basic weak classifier and the magnetic field strength weights of all magnetic field strength data during the iterative classification process; the classification error rate of the iterative weak classifier in the corresponding iterative classification process is used as the corresponding integration weight; a weighted sum is performed according to all the iterative weak classifiers and the corresponding integration weights to determine the human acupoint acupuncture model corresponding to each acupuncture point.

[0028] Further, the process of performing extremely weak magnetic measurement according to the human acupoint acupuncture model includes:

[0029] During the acupuncture process of the patient by integrated traditional Chinese and Western medicine acupuncture, the theoretical acupuncture angle and the theoretical acupuncture depth of the acupuncture point to be measured are input into the corresponding human acupoint acupuncture model to determine the theoretical magnetic field strength of the acupuncture point to be measured; calculate the extremely weak magnetic error representing the difference between the actually measured magnetic field strength data of the acupuncture point to be measured and the theoretical magnetic field strength.

[0030] When the extremely weak magnetic error is less than or equal to the preset allowable error, it is determined that the magnetic field strength data of the acupuncture point to be measured is normal data, and no abnormal measurement signal is sent.

[0031] When the extremely weak magnetic error is greater than the preset allowable error, it is determined that the magnetic field strength data of the acupuncture point to be measured is abnormal data, and an abnormal measurement signal is sent.

[0032] The present application also provides an extremely weak magnetic measurement system for an integrated traditional Chinese and Western medicine acupuncture department, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of any one of the extremely weak magnetic measurement methods for an integrated traditional Chinese and Western medicine acupuncture department are implemented.

[0033] The present application has the following beneficial effects:

[0034] This application first determines the classification error rate based on the classification error situation of the classifier during the iterative classification process, and screens out the correct magnetic field intensity data and incorrect magnetic field intensity data. Then, during the iterative adjustment of the weights of each magnetic field intensity data during the weak classifier training, based on the characteristic that the angle of needle insertion and the depth of needle insertion have a certain impact on the magnetic field change of the acupoint, a commonality analysis and a universality analysis are performed on the classified magnetic field intensity data at each needle insertion angle and each needle insertion depth, so as to determine the corresponding influencing factors, that is, the degree of influence. Thus, based on the influencing factors, the magnetic field intensity weights in the iterative classification process are adaptively adjusted, making the update of the weights during the weak classifier training more in line with objective facts, thereby improving the accuracy and convergence speed of the human acupoint acupuncture model constructed according to each weak classifier, and making the accuracy and efficiency of the extremely weak magnetic measurement higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0036] Figure 1 It is a flowchart of an extremely weak magnetic measurement method for a combined traditional Chinese and Western medicine acupuncture department provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of an extremely weak magnetic measurement method and system for a combined traditional Chinese and Western medicine acupuncture department proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as implying or indicating relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" can explicitly or implicitly include one or more of such features.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0039] The following specifically describes the specific solution of a very weak magnetic field measurement method and system for a traditional Chinese and Western medicine integrated acupuncture department provided by the present invention in conjunction with the accompanying drawings.

[0040] An embodiment of the present application provides a very weak magnetic field measurement method for a traditional Chinese and Western medicine integrated acupuncture department. Please refer to Figure 1 , which shows a flowchart of a very weak magnetic field measurement method for a traditional Chinese and Western medicine integrated acupuncture department provided by an embodiment of the present invention. The method includes:

[0041] Step S101: During the acupuncture treatment of patients in traditional Chinese and Western medicine integrated acupuncture, collect the magnetic field intensity data under each acupuncture point and the acupuncture angle and acupuncture depth when each magnetic field intensity data is collected.

[0042] In a specific implementation manner of the embodiment of the present invention, before acupuncture, monitor the magnetic field intensity curve of each acupuncture point of the patient during the acupuncture process through the arranged superconducting two-word interferometer, and use each peak value on the magnetic field intensity curve corresponding to each acupuncture point as all the magnetic field intensity data collected when acupuncture is performed on each acupuncture point; record the time point when each magnetic field intensity data is collected, and collect the corresponding acupuncture angle and acupuncture depth according to the corresponding time point.

[0043] In a specific implementation manner of the embodiment of the present invention, measure the external retention length of the acupuncture needle when each magnetic field intensity data under each acupuncture point is collected through a length measurement tool, and perform a negative correlation mapping on the ratio between the external retention length and the total length of the needle as the acupuncture depth collected in the embodiment of the present invention; among them, the acquisition process of the acupuncture depth is expressed by the formula: ; where is the acupuncture depth of the th acupuncture point when the th magnetic field intensity data is collected; is the external retention length of the acupuncture needle when the th magnetic field intensity data of the th acupuncture point is collected; is the corresponding total length of the needle.

[0044] In a specific implementation manner of the embodiment of the present invention, the process of obtaining the needle insertion angle includes: taking the connection line between two Yongquan acupoints corresponding to the soles of the patient's left and right feet as the horizontal axis, and taking the perpendicular line from the Baihui acupoint on the top of the head towards the horizontal axis as the vertical axis to establish a needle insertion angle analysis coordinate system; wherein, the Yongquan acupoint is located in the depression at the intersection of the anterior 1 / 3 and the posterior 2 / 3 of the connection line between the web margin of the 2nd and 3rd toes of the sole and the heel; the Baihui acupoint is located on the head, 5 cun directly above the midpoint of the anterior hairline; further, the included angle between the direction pointed by the needle tail of the inserted needle and the horizontal axis of the needle insertion angle analysis coordinate system is used as the collected needle insertion angle. It should be noted that the needle insertion angle and the needle insertion depth in the embodiment of the present invention are only used to distinguish the categories of magnetic field intensity data, that is, only exist as marks, and their actual sizes and dimensions have no influence on the specific implementation process of the present application. Therefore, the needle insertion angle and the needle insertion depth can be collected by different methods; for example, directly using the difference between the total length of the needle and the length retained outside the body as the needle insertion depth; using the included angle between the direction pointed by the needle tail or the needle tip of the inserted needle and the horizontal line as the needle insertion angle, which will not be further elaborated here; thus far, the magnetic field intensity data under each acupuncture point and the needle insertion angle and the needle insertion depth when each magnetic field intensity data is collected are obtained.

[0045] Step S102: Construct a corresponding basic weak classifier according to all the magnetic field intensity data under each acupuncture point; during the classification training process of the basic weak classifier, determine the correct magnetic field intensity data representing correct classification and the wrong magnetic field intensity data representing wrong classification corresponding to each iteration classification process of each acupuncture point; determine the classification error rate of each iteration classification process according to the number and weight distribution of the wrong magnetic field intensity data.

[0046] When using machine learning algorithms to establish a human acupoint acupuncture model, existing models generally only use acupoint differences and the magnetic field intensity corresponding to the acupoints for establishment. This establishment method is likely to lead to too high a misclassification rate and a slow convergence speed, and both the accuracy and the efficiency of the model are poor. When using acupuncture to stimulate acupoints, the insertion angle and the insertion depth of the needle have a certain impact on the magnetic field change of the acupoints. Therefore, on the basis of establishing the model, the present invention updates the model by using the specific behaviors during the acupuncture process on the basis of analyzing acupoint differences and the magnetic field intensity corresponding to the acupoints, so as to achieve a low misclassification rate and a high convergence speed. The bioelectric currents generated by acupuncture needles stimulating different acupoints are different, and thus the intensities of the generated magnetic fields may be different. And the final result required by the present invention is to perform magnetic measurement on each acupoint. Therefore, when establishing the basic classifier in the model, in order to avoid classification errors caused by cross-influence, it is necessary to classify the magnetic field intensity data, and then establish basic classifiers respectively. Therefore, the present application needs to establish and train basic classifiers for each acupuncture point to be inserted, so as to construct a corresponding human acupoint acupuncture model.

[0047] Preferably, in some possible implementation manners of the embodiments of the present invention, the process of obtaining the basic weak classifier includes:

[0048] Taking the reciprocal of the number of magnetic field intensity data under each acupuncture point as the initial weight of each magnetic field intensity data; based on the initial weight, constructing a decision tree for all magnetic field intensity data under each acupuncture point through a decision tree algorithm; and taking the decision tree as the basic weak classifier for each acupuncture point. In the process of establishing the base classifier in the prior art, the initial weights of all training samples are equal. Therefore, similarly, in this application, the reciprocal of the number of magnetic field intensity data is used as the initial weight of each magnetic field intensity data, so as to equally divide the weights of each magnetic field intensity data initially. It should be noted that the construction process of the basic weak classifier and the decision tree algorithm are common technical means for those skilled in the art, and will not be further limited and elaborated herein.

[0049] After constructing the basic weak classifier, it is further necessary to train the basic weak classifier, and construct a human acupoint acupuncture model through a weak classifier weighted integration method. Specifically, the training process is an iterative process, and the weights of each magnetic field intensity data are iterated during the classification process to obtain a new weak classifier. After the iteration ends, all weak classifiers are integrated to construct the final required human acupoint acupuncture model.

[0050] Furthermore, it is necessary to consider that the prior art usually corrects the magnetic field intensity weight by using the error of the weak classifier, and the error of the weak classifier is usually used as the weight for the integration of the final human acupoint acupuncture model; therefore, it is necessary to calculate the classification error rate for each iteration classification process during the classification training of the basic weak classifier. Preferably, in some possible implementation manners of the embodiments of the present invention, the process of obtaining the classification error rate includes:

[0051] Taking the weight of each magnetic field intensity data before each iteration classification process as the reference weight; and taking the ratio between the accumulated value of the reference weights of all correct magnetic field intensity data corresponding to each acupuncture point and the accumulated value of the reference weights of all magnetic field intensity data during each iteration classification process as the classification error rate of each acupuncture point during each iteration classification process.

[0052] Among them, the weight of each magnetic field strength data before each iteration classification process is the magnetic field strength weight calculated by the magnetic field strength data during the previous iteration classification process, which will not be elaborated further here; in the initial case, the weights of all magnetic field strength data are equal; that is, the reciprocal of the number of all magnetic field strength data in each acupuncture point is used as the weight of the magnetic field strength data before the first iteration classification process. If the classifier classifies all magnetic field strength data according to the decision tree, there is a possibility of classification error. Therefore, it is necessary to judge the classification effect of the weak classifier through the classification error, so as to adjust the weight of the subsequent magnetic field strength data according to the classification effect, making the classification result of the subsequent iteration classification process more accurate.

[0053] In a specific implementation manner of the embodiment of the present invention, the process of obtaining the classification error rate is expressed by the formula: ; where is the classification error rate of the th acupuncture point under the th iteration classification process; is the number of correct magnetic field strength data of the th acupuncture point under the th iteration classification process; is the total number of magnetic field strength data of the th acupuncture point; is the reference weight of the th correct magnetic field strength data of the th acupuncture point under the th iteration classification process; is the reference weight of the th magnetic field strength data of the th acupuncture point under the th iteration classification process. It should be noted that to ensure the meaningfulness of the calculation results, in the embodiment of the present invention, when performing fractional operations, in the case of encountering a denominator of 0, a tuning parameter factor greater than 0 needs to be added to the denominator to prevent the denominator from being 0. The value of the tuning parameter factor is set by the implementer according to the actual situation, and this application does not make special restrictions.

[0054] Step S103: Determine the data commonality of each acupuncture angle according to the fluctuation situation and the overall quantity of the correct magnetic field strength data corresponding to each acupuncture angle; determine the data universality of each acupuncture angle according to the overall quantity difference and the overall numerical difference between the incorrect magnetic field strength data and the correct magnetic field strength data corresponding to each acupuncture angle; determine the influence factor of each acupuncture angle according to the relative size between the data commonality and the data universality; determine the influence factor of each acupuncture depth.

[0055] The classification error rate of the weak classifier is obtained through step S102, and the classification error only depends on the magnetic field strength at the acupoint. As mentioned above, since different acupuncture angles and depths will affect the magnetic field strength, it is necessary to further analyze the magnetic field strength data obtained from the first classification for iterative weight adjustment. The commonality of the acupuncture behavior refers to whether there are obvious unified characteristics of the acupuncture angle and depth during the correct classification process; the unity refers to whether there is a certain similarity among different magnetic field strength data in the correct magnetic field strength data. If the correct magnetic field strength data at a certain acupuncture angle or depth are highly similar, it indicates that there is an obvious commonality in the correct magnetic field strength data for this acupuncture angle or depth, indirectly indicating that the acupuncture behavior at this acupuncture angle or depth will cause similar magnetic field strength data to be classified together, that is, the characteristics of this acupuncture angle or depth affect the fairness of the classification process. Therefore, the data commonality is further calculated based on the commonality characteristics. The data commonality represents the influence on the magnetic field strength data at the corresponding acupuncture angle or depth.

[0056] Preferably, in some possible implementation manners of the embodiment of the present invention, the process of obtaining the data commonality includes:

[0057] In each iterative classification process of each acupuncture point, the ratio between the number of correct magnetic field strength data corresponding to each acupuncture angle and the total number of correct magnetic field strength data is used as the reference correct proportion for each acupuncture angle; the product of the negative correlation mapping value of the variance of the correct magnetic field strength data corresponding to each acupuncture angle and the reference correct proportion is normalized to determine the data commonality corresponding to each acupuncture angle in each iterative classification process of each acupuncture point. According to the characteristics of the commonality, the closer the values of the correct magnetic field strength data for each acupuncture angle or depth are, the greater the commonality. Therefore, after performing a negative correlation mapping on the variance, the data commonality is calculated; and the quantity is used as a weight, that is, the more data for the corresponding acupuncture angle, the greater the proportion of this acupuncture angle. On this basis, the closer the values of the corresponding correct magnetic field strength data are, the greater the data commonality should be, and the greater the corresponding influence factor.

[0058] In a specific implementation manner of the embodiment of the present invention, the process of obtaining the data commonality of the acupuncture angle is represented by the formula: ; where is the data commonality corresponding to the acupuncture angle at the th iterative classification process of the th acupuncture point; is the rd acupuncture point at the Needling Angle in the Classification Process of the ith Iteration The number of corresponding correct magnetic field intensity data; For the th needling acupoint, the total number of correct magnetic field intensity data in the classification process of the th iteration; For the th needling acupoint, the reference correct proportion of the needling angle in the classification process of the th iteration. For the th needling acupoint, the variance of the correct magnetic field intensity data corresponding to the needling angle in the classification process of the th iteration. Is the exponential function with the natural constant as the base; Is the linear normalization function.

[0059] In a specific implementation manner of the embodiment of the present invention, the process of obtaining the data commonality of the needling depth is expressed by the formula: ; where, For the th needling acupoint, the data commonality corresponding to the needling depth in the classification process of the th iteration; For the th needling acupoint, the number of correct magnetic field intensity data corresponding to the needling depth in the classification process of the th iteration; For the th needling acupoint, the reference correct proportion of the needling depth in the classification process of the th iteration. For the th needling acupoint, the variance of the correct magnetic field intensity data corresponding to the needling depth in the classification process of the th iteration; the meanings of other parameters are the same as the formula explanations corresponding to the process of obtaining the data commonality of the needling angle, and will not be further elaborated; that is, the calculation methods of the data commonality of the needling depth and the needling angle are the same, and the subsequent data universality and influencing factors all conform to the characteristics of the same calculation method, so no further elaboration will be made hereinafter.

[0060] Further, the corresponding needle insertion behavior in the correct magnetic field intensity data needs to be analyzed jointly with the incorrect magnetic field intensity data. A greater data commonality indicates that the correct magnetic field intensity data exhibits a relatively close characteristic. If the characteristics of the incorrect magnetic field intensity data are relatively close to those of the normal magnetic field intensity data at this time, it means that the influence represented by the data commonality is not caused by the needle insertion behavior, that is, the needle insertion depth or the needle insertion angle. Therefore, on the basis of a large data commonality, if the correct magnetic field intensity data and the incorrect magnetic field intensity data have universality, it indicates that the process of classifying the incorrect magnetic field intensity data belongs to a normal classification process, and the corresponding needle insertion behavior has no significant impact; if universality is not available, it means that the needle insertion behavior has indeed affected the classification process, and some needle insertion behaviors need to be considered with emphasis. The magnetic field intensity data at the corresponding needle insertion angle or needle insertion depth has been severely affected, and the corresponding influence factor is large.

[0061] Preferably, in some possible implementation manners of the embodiments of the present invention, the process of obtaining data universality includes:

[0062] At each needle insertion angle, the difference between the mean value of all correct magnetic field intensity data and the mean value of all incorrect magnetic field intensity data is negatively correlated and mapped to determine the corresponding intensity universality; according to the difference between the total number of correct magnetic field intensity data and the total number of incorrect magnetic field intensity data, a negative correlation mapping is performed to determine the corresponding quantity universality; the product of the intensity universality and the quantity universality is normalized to determine the data universality corresponding to each needle insertion angle at each iterative classification process of each acupuncture point. Since there are a large number of different data in the incorrect magnetic field intensity data, the universality determined by using the mean value instead of the variance for the calculation method of universality is better; the closer the overall numerical size, that is, the mean value, between the corresponding correct magnetic field intensity data and the incorrect magnetic field intensity data, and the closer the quantity, the fairer the classification process, the more in line with the needle insertion behavior with universality, and the greater the corresponding data universality.

[0063] In a specific implementation manner of the embodiments of the present invention, the process of obtaining the data universality of the needle insertion angle is represented by the formula: ; where is the data universality corresponding to the needle insertion angle at the th iterative classification process of the th acupuncture point; is the number of correct magnetic field intensity data corresponding to the needle insertion angle at the th iterative classification process of the th acupuncture point; is the th acupuncture point of the Needling angle in the next iteration classification process The number of corresponding incorrect magnetic field strength data; For the th acupuncture point, the needling angle in the next iteration classification process The mean value of the corresponding correct magnetic field strength data; For the th acupuncture point, the needling angle in the next iteration classification process The mean value of the corresponding incorrect magnetic field strength data; Is the absolute value symbol; For the th acupuncture point, the needling angle in the next iteration classification process, and the corresponding intensity universality; For the th acupuncture point, the needling angle in the next iteration classification process, and the corresponding quantity universality; Is the linear normalization function.

[0064] Similarly, in a specific implementation manner of the embodiment of the present invention, the acquisition process of the data universality of the needling depth is expressed by the formula: ; where For the th acupuncture point, the needling depth in the next iteration classification process, and the corresponding data universality; For the th acupuncture point, the needling depth in the next iteration classification process, and the number of the corresponding correct magnetic field strength data; For the th acupuncture point, the needling depth in the next iteration classification process, and the number of the corresponding incorrect magnetic field strength data; For the th acupuncture point, the needling depth in the next iteration classification process, and the mean value of the corresponding correct magnetic field strength data; For the th acupuncture point, the needling depth in the next iteration classification process, and the mean value of the corresponding incorrect magnetic field strength data; Is the absolute value symbol; For the The needling depth under the corresponding intensity universality in the th iteration classification process for the th acupuncture point; The needling depth under the corresponding quantity universality in the th iteration classification process for the

[0065] The influence factor essentially characterizes the fairness or rationality of the corresponding classification process through the distribution characteristics of the magnetic field intensity data under the needling angle or needling depth. If the fairness or rationality indirectly reflected in the needling angle and needling depth of the magnetic field intensity data is worse, it indicates that the classification process is not fair enough and a greater weight needs to be based on the corresponding magnetic field intensity data. Therefore, the present application further needs to calculate the influence factor on the basis of data universality and data commonality, so as to indirectly measure the fairness of the classification of the magnetic field data under the corresponding needling angle or needling depth.

[0066] Preferably, in some possible implementation manners of the embodiments of the present invention, the process of obtaining the influence factor includes:

[0067] Determine the influence factor corresponding to each needling angle in each iteration classification process of each acupuncture point according to the product of data universality and data commonality. Since the greater the data commonality, the greater the corresponding data universality, the more significant the influence of the corresponding needling behavior, i.e., needling depth or needling angle, on the classification process. Therefore, by fusing data universality and data commonality through the product, the obtained influence factor is more accurate.

[0068] In a specific implementation manner of the embodiments of the present invention, the process of obtaining the influence factor of the needling angle is expressed by the formula: ; where is the influence factor corresponding to the needling angle in the th iteration classification process of the th acupuncture point; is the data commonality corresponding to the needling angle in the th iteration classification process of the th acupuncture point; is the data universality corresponding to the needling angle in the th iteration classification process of the th acupuncture point. Similarly, the process of obtaining the influence factor of the needling depth is expressed by the formula: ; where is the influence factor corresponding to the needling depth in the The needle insertion depth under the current iteration classification process corresponding influence factor; For the th needle insertion acupoint, the needle insertion depth under the current iteration classification process corresponding data commonality; For the th needle insertion acupoint, the needle insertion depth under the current iteration classification process corresponding data universality.

[0069] Step S104: In each iteration classification process, according to the classification error rate, the influence factors of the needle insertion angles corresponding to each magnetic field intensity data, and the influence factor of the needle insertion depth, determine the magnetic field intensity weight of each magnetic field intensity data after each iteration classification process; construct the human acupoint acupuncture model of each needle insertion acupoint according to the weak classifiers obtained during the magnetic field intensity weight iteration process; perform extremely weak magnetic measurement according to the human acupoint acupuncture model.

[0070] Preferably, in some possible implementation manners of the embodiments of the present invention, the process of obtaining the magnetic field intensity weight includes:

[0071] In each iteration classification process of each needle insertion acupoint, take the product of the normalized value of the classification error rate and the indication coefficient corresponding to each magnetic field intensity data as the initial weight update parameter for each magnetic field intensity data; wherein, the indication coefficient of the correct magnetic field intensity data is 1, and the indication coefficient of the incorrect magnetic field intensity data is 0; take the sum value between the influence factor of the needle insertion angle corresponding to each magnetic field intensity data and the influence factor of the needle insertion depth as the reference sum value; take the product of the negative correlation mapping value of the reference sum value and the initial weight update parameter as the correction weight update parameter; take the product of the negative correlation mapping value of the correction weight update parameter and the reference weight as the magnetic field intensity weight in each iteration classification process of each needle insertion acupoint.

[0072] For each magnetic field strength data during each iteration classification for each acupuncture point, the greater the influence factors of the acupuncture depth and the acupuncture angle corresponding to it, the greater the influence of the classification result of this magnetic field strength by the acupuncture depth and the acupuncture angle, and the more unfair the classification of this magnetic field strength data under the corresponding classification process. Therefore, greater attention needs to be paid in the next classification, and the corresponding magnetic field strength weight should be greater. The initial weight update parameter belongs to the existing parameters in the weight update process in the prior art, and it is obtained by the product of the misclassification rate after normalization and the indication coefficient, and its meaning will not be further elaborated here. Since the magnetic field strength weight is greater when the influence factor is greater, the reference sum value needs to be negatively correlated and mapped before calculating the corrected weight update parameter, and obtaining the magnetic field strength weight value according to the product of the negatively correlated mapped value of the weight update parameter and the reference weight also belongs to the conventional means of parameter weight update in the prior art, and will not be further elaborated here.

[0073] In a specific implementation manner of the embodiment of the present invention, the process of obtaining the magnetic field strength weight value is expressed by the formula: ; where is the magnetic field strength weight value of the th magnetic field strength data of the th acupuncture point after the th iteration classification process; is the reference weight of the th magnetic field strength data of the th acupuncture point in the th iteration classification process, that is, the magnetic field strength weight value after the previous iteration classification process; is the classification error rate of the th acupuncture point in the th iteration classification process; is the indication coefficient of the th magnetic field strength data of the th acupuncture point in the th iteration classification process; is the linear normalization function; is the influence factor of the acupuncture angle of the th magnetic field strength data of the th acupuncture point in the th iteration classification process; is the influence factor of the acupuncture depth of the th magnetic field strength data of the th acupuncture point in the th iteration classification process; is the th magnetic field strength data of the th acupuncture point in the Reference sum value under the iterative classification process; For the th acupuncture point, the th magnetic field intensity data, the initial weight update parameter under the th iterative classification process; For the th acupuncture point, the th magnetic field intensity data, the corrected weight update parameter under the th iterative classification process; Is the exponential function with the natural constant as the base. That is, compared with the weight adjustment formula in the prior art, only is added as the denominator to affect the weight update parameter.

[0074] After determining the magnetic field intensity weights corresponding to each magnetic field intensity data under each iterative classification process, further according to the specific process of the weak classifier weighted integration model in the prior art, further obtain the human acupoint acupuncture model, specifically including: under each acupuncture point, according to the basic weak classifier and the magnetic field intensity weights of all magnetic field intensity data in the iterative classification process, determine the iterative weak classifier corresponding to each iterative classification process; use the classification error rate of the iterative weak classifier in the corresponding iterative classification process as the corresponding integration weight; perform weighted summation according to all iterative weak classifiers and the corresponding integration weights to determine the human acupoint acupuncture model corresponding to each acupuncture point. It should be noted that the weak classifier weighted integration model is a well-known technical means in the art and will not be further limited and described here. In a specific implementation manner of the embodiment of the present invention, the stop condition of the iterative classification is set to iterate 10 times, and the iterative stop condition can be adjusted according to the specific implementation environment.

[0075] After constructing the weak classifier weighted integration model, further, a more accurate extremely weak magnetic field can be measured through the weak classifier weighted integration model. Preferably, the process of measuring the extremely weak magnetic field according to the human acupoint acupuncture model includes:

[0076] During the acupuncture treatment of patients by integrated traditional Chinese and Western medicine acupuncture, the theoretical acupuncture angle and theoretical acupuncture depth of the acupuncture point to be measured are input into the corresponding human acupuncture point model to determine the theoretical magnetic field intensity of the acupuncture point to be measured; calculate the extremely weak magnetic error that characterizes the difference between the actually measured magnetic field intensity data of the acupuncture point to be measured and the theoretical magnetic field intensity; when the extremely weak magnetic error is less than or equal to the preset allowable error, determine that the magnetic field intensity data of the acupuncture point to be measured is normal data and do not send an abnormal measurement signal; when the extremely weak magnetic error is greater than the preset allowable error, determine that the magnetic field intensity data of the acupuncture point to be measured is abnormal data and send an abnormal measurement signal. The theoretical acupuncture angle and theoretical acupuncture depth are the data that should be used for acupuncture in theory. Therefore, when performing acupuncture specifically, the corresponding acupuncture effect can be judged through the extremely weak magnetic error and corresponding adjustments can be made accordingly; in a specific implementation manner of the embodiment of the present invention, the preset allowable error is set to 0.5.

[0077] In summary, based on the characteristic that the acupuncture angle and acupuncture depth have a certain impact on the magnetic field change of the acupuncture point, the present method conducts a common analysis and a general analysis on the magnetic field intensity data classified under each acupuncture angle and each acupuncture depth, so as to determine the corresponding influencing factors, that is, the degree of influence; and then adaptively adjusts the magnetic field intensity weights in the iterative classification process based on the influencing factors, making the update of the weights in the weak classifier training process more in line with objective facts, thereby improving the accuracy and convergence speed of the human acupuncture point model constructed according to each weak classifier, and making the accuracy of the extremely weak magnetic measurement higher.

[0078] The present application also proposes an extremely weak magnetic measurement system for an integrated traditional Chinese and Western medicine acupuncture department, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the extremely weak magnetic measurement methods for an integrated traditional Chinese and Western medicine acupuncture department.

[0079] It should be noted that: the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0080] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An extremely weak magnetic measurement system for the integrated traditional Chinese and Western medicine acupuncture department, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the following is implemented: During the acupuncture treatment of patients with integrated traditional Chinese and Western medicine acupuncture, collect the magnetic field intensity data under each acupuncture point and the acupuncture angle and depth when each magnetic field intensity data is collected; According to all the magnetic field intensity data under each acupuncture point, construct the corresponding basic weak classifier; during the classification training of the basic weak classifier, determine the correct magnetic field intensity data representing correct classification and the incorrect magnetic field intensity data representing incorrect classification corresponding to each iteration classification process of each acupuncture point; determine the classification error rate of each iteration classification process according to the number and weight distribution of the incorrect magnetic field intensity data; According to the fluctuation situation and the overall quantity of the correct magnetic field intensity data corresponding to each acupuncture angle, determine the data commonality of each acupuncture angle; according to the overall quantity difference and the overall numerical difference between the incorrect magnetic field intensity data and the correct magnetic field intensity data corresponding to each acupuncture angle, determine the data universality of each acupuncture angle; determine the influence factor of each acupuncture angle according to the relative size between the data commonality and the data universality; determine the influence factor of each acupuncture depth; During each iteration classification process, determine the magnetic field intensity weight of each magnetic field intensity data after each iteration classification process according to the classification error rate, the influence factor of the acupuncture angle corresponding to each magnetic field intensity data, and the influence factor of the acupuncture depth; construct the human acupuncture point acupuncture model of each acupuncture point according to the weak classifiers obtained during the iteration process of the magnetic field intensity weight; perform extremely weak magnetic measurement according to the human acupuncture point acupuncture model.

2. The extremely weak magnetic measurement system for the integrated traditional Chinese and Western medicine acupuncture department according to claim 1, characterized in that, The process of obtaining the basic weak classifier includes: Take the reciprocal of the number of magnetic field intensity data under each acupuncture point as the initial weight of each magnetic field intensity data; based on the initial weight, construct a decision tree for all the magnetic field intensity data under each acupuncture point through the decision tree algorithm; take the decision tree as the basic weak classifier of each acupuncture point.

3. The extremely weak magnetic measurement system for the integrated traditional Chinese and Western medicine acupuncture department according to claim 1, characterized in that, The process of obtaining the classification error rate includes: Take the weight of each magnetic field intensity data before each iteration classification process as the reference weight; take the ratio between the cumulative value of the reference weights of all the correct magnetic field intensity data corresponding to each acupuncture point in each iteration classification process and the cumulative value of the reference weights of all the magnetic field intensity data as the classification error rate of each acupuncture point in each iteration classification process.

4. The extremely weak magnetic measurement system for the integrated traditional Chinese and Western medicine acupuncture department according to claim 1, characterized in that, The process of obtaining the data commonality includes: During each iteration classification process of each acupuncture point, take the ratio between the number of correct magnetic field intensity data corresponding to each acupuncture angle and the total number of correct magnetic field intensity data as the reference correct proportion of each acupuncture angle; Normalize the product of the negative correlation mapping value of the variance of the correct magnetic field intensity data corresponding to each acupuncture angle and the reference correct proportion to determine the data commonality corresponding to each acupuncture angle in each iteration classification process of each acupuncture point.

5. The extremely weak magnetic measurement system for the integrated traditional Chinese and Western medicine acupuncture department according to claim 1, wherein The process of obtaining the data universality includes: At each acupuncture angle, the difference between the mean of all correct magnetic field intensity data and the mean of all incorrect magnetic field intensity data is negatively correlated and mapped to determine the corresponding intensity universality; according to the difference between the total number of correct magnetic field intensity data and the total number of incorrect magnetic field intensity data, a negative correlation mapping is performed to determine the corresponding quantity universality; the product between the intensity universality and the quantity universality is normalized to determine the data universality corresponding to each acupuncture angle at each acupuncture point during each iterative classification process.

6. The extremely weak magnetic measurement system for integrated traditional Chinese and Western medicine acupuncture departments according to claim 1, characterized in that, The process for obtaining the influencing factor includes: According to the product between the data universality and the data commonality, determine the influencing factor corresponding to each acupuncture angle at each acupuncture point during each iterative classification process.

7. The extremely weak magnetic measurement system for integrated traditional Chinese and Western medicine acupuncture departments according to claim 3, characterized in that, The process for obtaining the magnetic field intensity weight includes: During each iterative classification process for each acupuncture point, the product of the normalized value of the classification error rate and the indication coefficient corresponding to each magnetic field intensity data is used as the initial weight update parameter for each magnetic field intensity data; among them, the indication coefficient of correct magnetic field intensity data is 1, and the indication coefficient of incorrect magnetic field intensity data is 0; The sum of the influencing factor of the acupuncture angle corresponding to each magnetic field intensity data and the influencing factor of the corresponding acupuncture depth is used as the reference sum value; the product of the negatively correlated mapped value of the reference sum value and the initial weight update parameter is used as the corrected weight update parameter; The product of the negatively correlated mapped value of the corrected weight update parameter and the reference weight is used as the magnetic field intensity weight for each acupuncture point during each iterative classification process.

8. The extremely weak magnetic measurement system for the integrated traditional Chinese and Western medicine acupuncture department according to claim 1, characterized in that, The process for obtaining the human acupoint acupuncture model includes: At each acupuncture point, according to the basic weak classifier and the magnetic field intensity weights of all magnetic field intensity data during the iterative classification process, determine the iterative weak classifier corresponding to each iterative classification process; use the classification error rate of the iterative weak classifier during the corresponding iterative classification process as the corresponding integrated weight; perform weighted summation according to all iterative weak classifiers and the corresponding integrated weights to determine the human acupoint acupuncture model corresponding to each acupuncture point.

9. The extremely weak magnetic measurement system for integrated traditional Chinese and Western medicine acupuncture departments according to claim 1, wherein, The process for performing extremely weak magnetic measurement according to the human acupoint acupuncture model includes: During the acupuncture process of the patient by integrated traditional Chinese and Western medicine acupuncture, input the theoretical acupuncture angle and theoretical acupuncture depth of the acupuncture point to be measured into the corresponding human acupoint acupuncture model to determine the theoretical magnetic field intensity of the acupuncture point to be measured; calculate the extremely weak magnetic error representing the difference between the actually measured magnetic field intensity data of the acupuncture point to be measured and the theoretical magnetic field intensity; When the extremely weak magnetic error is less than or equal to the preset allowable error, determine that the magnetic field intensity data of the acupuncture point to be measured is normal data and no abnormal measurement signal is sent; When the extremely weak magnetic error is greater than the preset allowable error, determine that the magnetic field intensity data of the acupuncture point to be measured is abnormal data and an abnormal measurement signal is sent.

Images