Depth controllable piezoelectric shock wave therapy system based on automatic focusing

Abstract

The present application relates to the technical field of medical instrument control, in particular to a depth controllable piezoelectric shock wave treatment system based on automatic focusing. A data acquisition module system collects sample data of patients with the same disease type, including historical and to-be-tested samples, covering multi-dimensional information such as treatment feedback text and treatment area characteristic parameters. Due to individual differences among patients, a sample data analysis module obtains sample tolerance sensitivity by analyzing pain test data, determines the influence degree of complications by comparing treatment feedback texts, and accurately grasps the physiological and pathological characteristics of individual patients. A linkage characteristic analysis module comprehensively considers similar characteristics such as treatment area characteristic parameters, determines the linkage characteristic performance degree among samples, and excavates the internal relationship and similar rules. A treatment control parameter determination module trains a neural network based on the linkage characteristic performance degree and treatment control parameters, accurately determines the treatment control parameters of the to-be-tested sample by using the learning and prediction ability of the neural network, and avoids over-treatment or under-treatment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical instrument control, in particular to a depth-controllable piezoelectric shock wave treatment system based on automatic focusing. BACKGROUND

[0002] Piezoelectric shock wave treatment technology, especially depth-controllable piezoelectric shock wave, as a non-invasive treatment method, is widely used in the medical field, such as orthopedic rehabilitation, chronic pain management, etc. Piezoelectric shock wave treatment technology can effectively stimulate blood circulation, etc. by acting on human tissues with high-energy sound waves. The depth-controllable piezoelectric shock wave treatment system combines precise focusing technology (the core of focusing is to control the focusing position of energy waves to ensure that energy is accurately applied to the target area), which can adjust various treatment control parameters of piezoelectric shock wave according to different diseases to improve the treatment effect.

[0003] In the prior art, when setting various treatment control parameters of piezoelectric shock wave during the treatment of patients, the treatment control parameters of historical samples with similar treatment area characteristics are often selected as a reference. However, since each patient has its unique physiological and pathological characteristics, the pain tolerance and treatment effect vary from person to person. Therefore, if only the similar situation of the treatment area characteristics is relied on and the individual differences of patients are ignored, the setting accuracy of the treatment control parameters of the test sample will be low, which may seriously affect the treatment effect. SUMMARY

[0004] In order to solve the technical problem that the setting accuracy of the treatment control parameters of the test sample will be low if only the similar situation of the treatment area characteristics is relied on and the individual differences of patients are ignored, the present application aims to provide a depth-controllable piezoelectric shock wave treatment system based on automatic focusing, and the technical solution is as follows:

[0005] The present application provides a depth-controllable piezoelectric shock wave treatment system based on automatic focusing, which comprises:

[0006] A data acquisition module is configured to acquire sample data of patients with the same disease type, wherein the sample data includes historical samples and test samples, and each sample data includes treatment feedback text, treatment area characteristic parameters, treatment control parameters, and pain test data of the patient at each treatment.

[0007] a sample data analysis module, configured to analyze a change trend of the pain test data and a similar feature of a numerical value in the sample data, and obtain a tolerance sensitivity of each sample data; and determine a complication influence degree of each sample data by comparing similarities between the treatment feedback texts in each sample data;

[0008] a linkage feature analysis module, configured to comprehensively determine a linkage feature performance degree between any two sample data based on similar features of the treatment area feature parameters, similar features of the tolerance sensitivities, and similar features of the complication influence degrees between the sample data;

[0009] a treatment control parameter determination module, configured to determine a treatment control parameter of a to-be-tested sample based on the linkage feature performance degrees between the sample data and the treatment control parameter, and train a neural network.

[0010] Further, the tolerance sensitivity acquisition method comprises:

[0011] The pain test data comprises subjective pain scores under each test energy and electromyographic signal data, and the test energy is gradually increased;

[0012] In each sample data, a subjective pain sensitivity of each sample data is determined according to a change trend of the subjective pain scores;

[0013] Similar features and numerical features of the electromyographic signals between the sample data are analyzed to determine an objective pain sensitivity of each sample data;

[0014] A value obtained by multiplying the subjective pain sensitivity and the objective pain sensitivity of each sample data after normalization is taken as the tolerance sensitivity of each sample data.

[0015] Further, the subjective pain sensitivity acquisition method comprises:

[0016] In each sample data, the subjective pain scores are sorted according to an increasing order of the test energy to obtain a sorting sequence;

[0017] In the sorting sequence, a ratio of each subjective pain score to an adjacent previous pain score is calculated as a pain increase perception degree of each subjective pain score;

[0018] In each sample data, the subjective pain scores are weighted and fused by using the pain increase perception degrees of each subjective pain score, and a value obtained by normalizing the weighted result is taken as the subjective pain sensitivity of each sample data.

[0019] Further, the objective pain sensitivity acquisition method comprises:

[0020] In the pain test data of each sample data, the electromyographic signal data under each test energy is taken as the electromyographic energy representation value of the absolute area of the horizontal axis;

[0021] For any two sample data, the difference characteristics of the electromyographic energy representation values of the two sample data under the same test energy are analyzed, and the objective stimulation performance difference degree between the two sample data is obtained in combination with the test energy numerical characteristics;

[0022] Based on the objective stimulation performance difference degree between the sample data, cluster analysis is performed on all sample data, so as to obtain all cluster clusters;

[0023] In each cluster cluster, the normalized value of the mean value of the peak values of the electromyographic signals of all sample data under all test energies is taken as the objective pain sensitivity of each sample data in each cluster cluster.

[0024] Further, the objective stimulation performance difference degree acquisition method comprises:

[0025] For any two sample data, the absolute value of the difference value of the electromyographic energy representation values of the two sample data under the same test energy is taken as the electromyographic difference characteristic value of the two sample data under the same test energy;

[0026] The ratio of each test energy to the sum of all test energy values is taken as the stimulation intensity weight of each test energy;

[0027] For any two sample data, the electromyographic difference characteristic value of the two sample data under the same test energy is weighted and fused by using the stimulation intensity weight of the test energy, and the normalized value of the obtained weighted result is taken as the objective stimulation performance difference degree of the two sample data.

[0028] Further, the complication influence degree acquisition method comprises:

[0029] The complication keyword set of each sample data at each treatment is extracted from the treatment feedback text at each treatment, so as to obtain the complication keyword set at each treatment;

[0030] In each sample data, the Jaccard correlation coefficient between the complication keyword set at the last treatment and the complication keyword set at each previous treatment is calculated as a complication similarity factor;

[0031] The sum of all complication similarity factors corresponding to each sample data is negatively correlated and mapped, and the value after the mapping is taken as the complication influence degree of each sample data.

[0032] Further, the linkage feature performance degree acquisition method comprises:

[0033] The similarity of the treatment region between the two sample data is obtained by analyzing the similar features of the treatment region feature parameters between any two sample data.

[0034] The value obtained by negatively correlating and normalizing the absolute value of the difference of the tolerance sensitivity between the two sample data is taken as the tolerance sensitivity similarity between the two sample data.

[0035] The value obtained by negatively correlating and normalizing the absolute value of the difference of the complication influence degree between the two sample data is taken as the complication influence similarity between the two sample data.

[0036] The linkage feature performance degree between the two sample data is obtained by weighting the tolerance sensitivity similarity with the treatment region similarity, weighting the complication influence similarity with the constant 1 minus the difference of the treatment region similarity, and normalizing the weighted result.

[0037] Further, the treatment region similarity acquisition method comprises:

[0038] The treatment region feature parameters at least include the skin-to-target distance and the subcutaneous fat layer thickness, and the treatment region feature parameters of each sample data at each treatment are combined to form a feature parameter vector.

[0039] For any two sample data, the Euclidean distance between any two feature parameter vectors of the two sample data is calculated, negatively correlated and normalized to obtain the treatment region similarity factor between the two sample data, and the mean value of all treatment region similarity factors between the two sample data is taken as the treatment region similarity.

[0040] Further, the neural network is trained based on the linkage feature performance degree between the sample data and the treatment control parameter, so as to determine the treatment control parameter of the to-be-tested sample, which comprises:

[0041] Based on the linkage feature performance degree between all historical samples, the clustering analysis is performed on all historical samples to obtain a clustering cluster.

[0042] In each clustering cluster, the mean value of the linkage feature performance degree between each historical sample and other historical samples is taken as the treatment performance feature value of each historical sample.

[0043] In each clustering cluster, the treatment performance feature value of each historical sample is taken as the input of the neural network, the mean value of each historical sample under each treatment control parameter is calculated as the control parameter factor, and the control parameter factors of each historical sample under all kinds of treatment control parameters are taken as the output, so as to train the neural network to obtain the trained neural network.

[0044] The to-be-tested sample is divided into the cluster to which the historical sample corresponding to the maximum linkage feature performance degree of the to-be-tested sample belongs, and the average linkage feature performance degree between the to-be-tested sample and all the historical samples in the cluster to which the to-be-tested sample belongs is taken as the input of the trained neural network corresponding to the cluster, so that the value of each treatment control parameter corresponding to the to-be-tested sample is obtained.

[0045] Further, the clustering analysis of all the historical samples based on the linkage feature performance degrees between all the historical samples is performed to obtain the cluster, including:

[0046] In all the historical samples, the clustering analysis of all the historical sample data is performed based on the K-means clustering algorithm and the preset optimal K value, to obtain all the clusters, wherein the distance measurement is the value after the negative correlation mapping of the linkage feature performance degree between the historical sample data.

[0047] The application has the following beneficial effects:

[0048] The data acquisition module can systematically collect patient sample data of the same disease type, and the sample data is divided into historical samples and to-be-tested samples, and covers multidimensional information such as treatment feedback text, treatment area feature parameters, treatment control parameters and pain test data. Since patients have individual differences, the pain tolerance and feedback after treatment of each person are different, therefore, in the sample data analysis module, the tolerance sensitivity of each sample data is accurately obtained by analyzing the change trend and similar characteristics of the value of the pain test data in the sample data; meanwhile, the complication influence degree is accurately determined by comparing the similarity of the treatment feedback text of the sample data; this module can accurately grasp the individual physiological and pathological characteristics of the patient, and provides a key basis for personalized treatment, which helps to improve the pertinence and effectiveness of treatment. Further, in the linkage feature analysis module, the similarity characteristics of the treatment area feature parameters, the tolerance sensitivity and the complication influence degree between the sample data are comprehensively considered, to determine the linkage feature performance degree between any two sample data. Thus, by mining the internal relationship and similar rules between different sample data, it is helpful to more reasonably refer to the historical samples in the subsequent process, to provide a more scientific reference for the setting of the treatment control parameters of the to-be-tested sample. Therefore, finally, in the treatment control parameter determination module, the neural network is trained based on the linkage feature performance degree between the sample data and the treatment control parameters, and the strong learning and prediction ability of the neural network is utilized, to accurately determine the treatment control parameters of the to-be-tested sample according to the individual characteristics of the patient and the linkage relationship between the sample data. In summary, the application can fully consider the individual differences and similar characteristics between the sample data, and can accurately set the treatment control parameters, to effectively avoid the occurrence of over-treatment or under-treatment. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 is a system block diagram of a depth controllable piezoelectric shock wave treatment system based on automatic focusing provided by an embodiment of the present application;

[0051] Figure 2 is a method flow chart of a sensitivity tolerance acquisition method provided by an embodiment of the present application;

[0052] Figure 3 is a method flow chart of a linkage feature performance degree acquisition method provided by an embodiment of the present application;

[0053] Figure 4 is a system structure schematic diagram of a depth controllable piezoelectric shock wave treatment system based on automatic focusing provided by an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined invention purpose, the following combines the drawings and preferred embodiments to specifically describe a depth controllable piezoelectric shock wave treatment system based on automatic focusing provided by the present application, its specific implementation, structure, features and effects, which are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0056] The following specifically describes a specific scheme of a depth controllable piezoelectric shock wave treatment system based on automatic focusing provided by the present application in combination with the drawings.

[0057] Please refer to Figure 1 , which shows a system block diagram of a depth controllable piezoelectric shock wave treatment system based on automatic focusing provided by an embodiment of the present application. The system comprises a data acquisition module 101, a sample data analysis module 102, a linkage feature analysis module 103, and a treatment control parameter determination module 104.

[0058] The data acquisition module 101 is configured to acquire sample data of patients with the same disease type, the sample data including historical sample data and to-be-tested sample data, and each sample data including treatment feedback text, treatment area feature parameters, treatment control parameters of the patient at each treatment time, and pain test data of the patient.

[0059] When performing piezoelectric shock wave treatment, the system integrates a real-time imaging monitoring (such as B-ultrasound or X-ray), and through locking a target (such as a stone), the computer system automatically calculates the accurate depth and position of the target, and automatically controls the piezoelectric ceramic array to dynamically change the focus position of the shock wave convergence through complex electronic phase control technology, so that the focus position is always automatically aligned and tracked to the moving target (such as a tumor in the kidney or liver that moves slightly when breathing).

[0060] However, this efficient real-time treatment system also depends on the preoperative multidimensional comprehensive analysis of the patient's disease. By comprehensively considering the patient's physical condition, lesion site, target size and shape, and other factors, the doctor can better adjust the treatment control parameters during treatment, such as the frequency of the shock wave, the number of shocks, the energy intensity, and the treatment time, and the optimization of these treatment control parameters can improve the treatment effect, reduce side effects, and provide protection for personalized treatment.

[0061] In this embodiment of the application, the data acquisition module can extract sample data of patients with the same disease type from the hospital database, such as kidney stones or liver tumors; wherein the sample data of the historical patients is denoted as historical sample data, and the sample data of the to-be-tested patients is denoted as to-be-tested sample data.

[0062] Because the patient does not perform piezoelectric shock wave treatment at one time, but considers the patient's condition and discomfort of single treatment for staging and multiple treatments, the treatment feedback text can be extracted from the medical record after each treatment of the patient, which is used to evaluate the patient's feeling and possible complications at each treatment time; at the same time, the spatial geometric structure of the treatment area determines the shape, position and size of the lesion tissue, and the treatment control parameters reflect the treatment plan, which is also the main mark of the patient's individual differences, so the treatment area feature parameters and the treatment control parameters are also extracted from the medical record after each treatment.

[0063] At the same time, because different patients have different pain tolerance, in order to ensure the comfort and effectiveness of the patient's treatment, the patient's pain tolerance must be considered, so in this embodiment of the application, the pain test is performed on each patient, and the pain test data of each patient is obtained.

[0064] At this point, multiple historical sample data and to-be-tested sample data under the same disease type can be obtained.

[0065] It should be noted that in this embodiment of the present application, the treatment area characteristic parameters at least include the skin-to-target distance and the subcutaneous fat layer thickness, etc., and the treatment control parameters at least include the shock wave frequency, the shock number, the energy intensity and the treatment time, etc.

[0066] Pain test process: Test area: normal tissue near the planned treatment target.

[0067] Test parameters:

[0068] (1) Fixed reference test energy: use an absolutely safe reference energy much lower than the treatment level.

[0069] (2) Incremental energy / frequency: emit a series of (e.g., 3-5 times) shock waves, and the energy is slightly increased according to the preset gradient (e.g., 0.2 times the original basis each time).

[0070] (3) Fixed focal depth: placed at a certain depth (e.g., 1 cm) under the skin.

[0071] In this way, the patient performs a pain performance score (0-10 points, with the pain level gradually increasing as the score increases, and no score is given after the first pulse, which is uniformly defaulted to 1 point, so that the patient can use it as a reference for comparison after subsequent pulses) after each pulse, which is used as the patient's subjective pain score. At the same time, through the attached surface electromyography (sEMG) sensor, the electromyography data after each pulse is obtained. Based on the foregoing steps, the pain test data corresponding to each sample data is obtained, and the pain test data includes the subjective pain score and the electromyography data under each test energy.

[0072] In the embodiment of the present application, the collection and acquisition of patient personal information data are all authorized by relevant users, and the process does not violate relevant laws and regulations and does not violate public order and good customs.

[0073] The sample data analysis module 102 is configured to analyze the change trend and the similar characteristics of the numerical values of the pain test data in the sample data, and obtain the tolerance sensitivity of each sample data. In each sample data, the similarity between the treatment feedback texts is compared to determine the complication influence degree of each sample data.

[0074] In actual treatment process, individual differences of patients, especially pain threshold and tolerance of patients to treatment can be quite different, which depends not only on physiological conditions of patients, but also can be affected by psychological factors. Thus, in order to ensure comfort and effectiveness of subsequent patient treatment, it is necessary to better understand the pain sensitivity difference of patients, so as to analyze the change trend and similar characteristics of pain test data in sample data, calculate the tolerance sensitivity of each sample data, and use the tolerance sensitivity to represent the pain tolerance of the patient corresponding to each sample data.

[0075] Preferably, the method for obtaining the tolerance sensitivity in an embodiment of the present application comprises:

[0076] Referring to Figure 2 , a method flowchart of the method for obtaining the tolerance sensitivity in an embodiment of the present application is shown, which comprises the following steps:

[0077] Step S201: In each sample data, the subjective pain sensitivity of each sample data is determined according to the change trend of the subjective pain score.

[0078] The test energy is a key factor affecting the pain perception of patients, and as the test energy increases, the pain reflection of patients usually also increases, so in each sample data, the subjective pain score is sorted according to the increasing order of the test energy, and a sorted sequence is obtained.

[0079] Then in the sorted sequence, the ratio of each subjective pain score to the adjacent previous pain score is calculated as the pain perception of each subjective pain score, which can measure the relative change degree of the pain degree under the adjacent two test energies, and the greater the index is, the greater the pain sensitivity of the patient is compared with the previous test energy.

[0080] Finally, in each sample data, the pain perception of each subjective pain score is weighted and fused: after the pain perception is normalized, it is used as the pain perception weight, the product of the pain perception weight under each test energy and the subjective pain score is used as the weighted score, and the sum of all weighted scores corresponding to each sample data is normalized, and the value after normalization is used as the subjective pain sensitivity of each sample data. The greater the subjective pain sensitivity is, the greater the pain sensitivity of the patient corresponding to the sample data is. The normalization is a technology known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited herein; the sum of the pain perception weights in each sample data is 1.

[0081] It should be noted that for the first subjective pain score in the sequence, since it does not have a previous subjective pain score, the pain perception of the first subjective pain score is set to be consistent with the second subjective pain score.

[0082] Step S202: Analyze the similar features and numerical features of the electromyographic signals between the sample data, and determine the objective pain sensitivity of each sample data.

[0083] The electromyographic signal is a direct physiological manifestation of muscle activity, and when stimulated by pain, the muscle will produce electrical activity. Therefore, in the pain test data of each sample data, the electromyographic signal data under each test energy is taken as the electromyographic energy representation value with the absolute area of the horizontal axis. The absolute area can be calculated by integrating the electromyographic signal data to obtain the integral value of the electromyographic signal data. The essence is to calculate the area between the signal data curve and the horizontal axis. The area in the region above the horizontal axis (vertical coordinate greater than 0) is positive, and the area in the region below the horizontal axis (vertical coordinate less than 0) is negative. Therefore, the absolute area here is the area in the region above the horizontal axis minus the area in the region below the horizontal axis.

[0084] In the pain test, the increase in test energy intensity is usually helpful to reveal the differences in stimulation perception between patients, especially when the intensity is greater, the individual's perception of stimulation will be more obvious, especially in the response of the nervous system. For patients with some diseases or physiological conditions, stronger stimulation can trigger more significant physiological responses, so that the objective pain tolerance of different patients can be better evaluated.

[0085] Therefore, for any two sample data, the absolute value of the difference between the two sample data is taken as the muscle energy difference characteristic value of the two sample data under the same test energy. The greater the muscle energy difference characteristic value, the greater the difference between the pain perception tolerance of the two sample data. Then, the ratio of each test energy to the sum of all test energies is taken as the stimulation intensity weight of each test energy. For any two sample data, the muscle energy difference characteristic value of the two sample data under the same test energy is weighted and fused by using the stimulation intensity weight of the test energy, that is, the stimulation intensity weight of each test energy is multiplied by the muscle energy difference characteristic value of the two sample data under each test energy to obtain a weighted difference characteristic value. The greater the value, the lower the similarity of the two sample data in the perception of pain under greater energy stimulation. Then, the sum of the weighted difference characteristic values of the two sample data under all test energies is normalized to obtain the objective stimulation performance difference of the two sample data. The greater the objective stimulation performance difference, the greater the pain perception degree of the two sample data under the same energy test. The normalization is a technical means familiar to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited herein.

[0086] Based on the objective stimulation performance difference between the sample data, all sample data are subjected to cluster analysis to obtain all cluster clusters. The algorithm of the cluster analysis can use the K-means algorithm, and the distance measurement is the objective stimulation performance difference between the sample data. The preset K value is 5, and the specific value can be adjusted according to the implementation scene, which is not limited herein.

[0087] At this time, the sample data in each cluster cluster has a relatively consistent pain stimulation perception degree. Since the peak value of the electromyographic signal represents the maximum intensity of muscle contraction, and the more intense the pain degree, the more intense the muscle contraction, finally, in each cluster cluster, the normalized value of the mean value of the peak values of the electromyographic signals of all sample data under all test energies is taken as the objective pain sensitivity of each sample data in each cluster cluster. The greater the objective pain sensitivity, the more obvious the pain reflection represented by the electromyographic signal of the patient. The normalization is a technical means familiar to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited herein.

[0088] Step S203: The subjective pain sensitivity and the objective pain sensitivity of each sample data are fused to obtain the tolerance sensitivity of each sample data.

[0089] Based on the analysis in step S201 and step S202, the subjective pain sensitivity and the objective pain sensitivity of each sample data can be calculated, and both of the two indicators are positively correlated with the response sensitivity of the patient to pain, so the product of the subjective pain sensitivity and the objective pain sensitivity of each sample data is normalized as the tolerance sensitivity of each sample data, and the greater the tolerance sensitivity, the more obvious the pain perception of the sample data in the treatment process, and the index can be used as a key indicator reflecting the individual differences of the sample data. The normalization is a technical means familiar to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited here.

[0090] Further, since the physical condition, underlying disease and other factors of each patient also differ, these factors will affect the occurrence and severity of complications, so the similarity between the treatment feedback texts of each sample data can be compared to determine the complication influence degree of each sample data, which is also a key indicator reflecting the individual differences of the sample data.

[0091] Preferably, the method for obtaining the complication influence degree in an embodiment of the present application comprises:

[0092] Based on the medical data, the complication keywords of each sample data at each treatment are extracted to obtain the complication keyword set at each treatment.

[0093] In each sample data, the Jaccard correlation coefficient between the complication keyword set at the last treatment and the complication keyword set at each previous treatment is calculated as a complication similarity factor, and the greater the complication similarity factor, the higher the similarity of the complication keyword sets at the two treatments, and the more similar the complication performance, and the reaction is more likely to conform to the common reaction mode of the patient and be within the expected range, and the treatment plan does not cause new abnormal conditions and belongs to the normal treatment reaction range, and vice versa, if the complication similarity factor is smaller, it means that the complication performance after the two treatments is quite different, and the last treatment may have an inappropriate impact on the patient's tissue or physiological state.

[0094] Therefore, the sum of all complication similarity factors corresponding to each sample data is subjected to negative correlation mapping processing to correct the logical relationship, and the complication influence degree of each sample data is obtained, and the greater the complication influence degree, the greater the probability of accidental tissue damage and other complications that the sample data corresponding to the patient may suffer during the treatment process, and the greater the degree of influence of the complications. The negative correlation mapping here can adopt the formula wherein, represents the exponential function with the natural constant e as the base, and x represents the independent variable.

[0095] The linkage feature analysis module 103 is configured to comprehensively analyze the similarity of the treatment area feature parameters, the similarity of the tolerance sensitivity, and the similarity of the complication influence degree between the sample data, and determine the linkage feature performance degree between any two sample data.

[0096] The treatment area feature parameters of the sample reflect the key information such as the lesion site and the treatment range, so the similarity of the treatment area feature parameters is taken into account for analysis, which can ensure that the difference in the treatment site on the treatment result is fully considered when the sample linkage is considered. The tolerance sensitivity of the sample data has individual differences, so the similarity of the tolerance sensitivity between the sample data can reflect the difference in the reaction of the patient when receiving treatment. At the same time, the complication is an important factor that cannot be ignored in the treatment process, which will affect the rehabilitation process and the treatment effect of the patient, and the similarity of the complication influence degree between the sample data can reflect the similarity of the risk and severity of the complication of different patients in the treatment process. The similarity of the three dimensions is related to and influences each other, so in the embodiment of the present application, the similarity of the three dimensions is comprehensively considered, which can more comprehensively and accurately determine the linkage feature performance degree between the sample data, and provide a more reliable reference basis for the subsequent analysis process.

[0097] Preferably, the method for obtaining the linkage feature performance degree in the embodiment of the present application comprises the following steps:

[0098] Please refer to Figure 3 which shows the method flow chart of the method for obtaining the linkage feature performance degree in the embodiment of the present application, and the method comprises the following steps:

[0099] Step S301: analyzing the similarity of the treatment area feature parameters between any two sample data to obtain the treatment area similarity between the two sample data.

[0100] The treatment area feature parameters at least include the skin-to-target distance and the subcutaneous fat layer thickness, and the treatment area feature parameters of each sample data at each treatment are combined to form a feature parameter vector.

[0101] The similarity of the treatment region characteristic parameters between the sample data determines whether the two sample data maintain similar relevant structural characteristics of the treatment region, and therefore, for any two sample data, the Euclidean distance between any two characteristic parameter vectors of the two sample data is calculated. The smaller the Euclidean distance, the more similar the two sample data are. Therefore, the Euclidean distance is negatively correlated and normalized to correct the logical relationship and obtain the treatment region similarity factor between the two sample data. At this time, the two sample data have a treatment region similarity factor between each two treatments, and the greater the value, the higher the similarity of the structural characteristics of the treatment region between the two sample data. Therefore, the mean of all treatment region similarity factors between the two sample data is finally taken as the treatment region similarity. The negative correlation and normalization can be performed by the formula wherein, represents an exponential function with a natural constant e as the base, and x represents the independent variable.

[0102] It should be noted that, for example, sample data A has 3 treatment records, denoted as treatment 1, treatment 2 and treatment 3, and sample data B has 2 treatment records, denoted as treatment 4 and treatment 5. There are 6 treatment region similarity factors between sample data A and sample data B, i.e., one treatment region similarity factor between treatment 1 and treatment 4, one treatment region similarity factor between treatment 1 and treatment 5, one treatment region similarity factor between treatment 2 and treatment 4, one treatment region similarity factor between treatment 2 and treatment 5, one treatment region similarity factor between treatment 4 and treatment 4, and one treatment region similarity factor between treatment 3 and treatment 5.

[0103] Step S302: Analyze the similar characteristics of the tolerance sensitivity between any two sample data, and determine the tolerance sensitivity similarity between the two sample data.

[0104] The tolerance sensitivity of the sample data reflects the pain perception of the patient corresponding to the sample data during the treatment process. This index can be used as a key index to reflect the individual differences of the sample data. Different patients have different tolerance abilities to treatment, which affects the development of treatment plan and treatment effect.

[0105] Therefore, for any two sample data, the absolute value of the difference between the tolerance sensitivity of the two sample data is calculated. The smaller the absolute value of the difference, the higher the similarity of the two sample data in terms of tolerance sensitivity. Therefore, the absolute value of the difference is negatively correlated and normalized to obtain the tolerance sensitivity similarity between the two sample data. The higher the tolerance sensitivity similarity, the more consistent the physiological state and pain perception ability of the patients corresponding to the two sample data during the treatment of the piezoelectric shock wave. The negative correlation and normalization can be performed by the formula wherein, denotes an exponential function with the natural constant e as the base, and x denotes an argument.

[0106] Step S303: Analyzing the similar features of the complication influence degree between any two sample data, and determining the complication influence similarity between the two sample data.

[0107] The complication influence degree of sample data indicates the probability of the sample data corresponding patient possibly suffering from complications such as accidental tissue damage during the treatment process, and the degree of influence of the complications, which is also a key indicator reflecting the individual differences of sample data. The complication influence degree of different patients is affected by various factors, such as treatment methods, underlying diseases, physical conditions, etc., so the differences in the risk and severity of complications that patients may produce during the treatment process will also affect the formulation of the treatment plan and the treatment effect.

[0108] Therefore, for any two sample data, the absolute value of the difference between the complication influence degrees of the two sample data is calculated. The smaller the absolute value, the higher the consistent features between the complication influence degrees of the two sample data, so the absolute value is negatively correlated and normalized to correct the logical relationship, and the complication influence similarity between the two sample data is obtained. The greater the complication influence similarity, the higher the consistent features of abnormal physiological responses and tissue damage degrees reflected by the two sample data during the treatment process. The negative correlation mapping and normalization processing here can use the formula wherein, denotes an exponential function with the natural constant e as the base, and x denotes an argument.

[0109] Step S304: Using the treatment area similarity between any two sample data to weight and fuse the tolerance sensitivity similarity and the complication influence similarity, thereby obtaining the linkage feature performance degree between the two sample data.

[0110] Based on the analysis in the foregoing steps, the treatment interval similarity, the tolerance sensitivity similarity, and the complication influence similarity between any two sample data can be obtained. The treatment interval similarity reflects the similarity of the structural features of the treatment areas of the two sample data. The greater the value, the more attention needs to be paid to the similar features of the physiological state and pain perception of the patients represented by the two sample data under piezoelectric shock wave treatment, so as to reasonably grasp the consistent direction of the overall treatment control parameters. Conversely, if the treatment interval similarity is smaller, it means that the structural features of the treatment areas of the two sample data are less similar, and therefore more detailed evaluation of the similar features of the complication risk of the patients represented by the sample data after treatment is needed.

[0111] Therefore, in this case, the treatment area similarity between the two sample data is used to weight the tolerance sensitivity similarity (the treatment area similarity and the tolerance sensitivity similarity are multiplied), and the difference value of the treatment area similarity is used to weight the complication influence similarity (the difference value of the treatment area similarity is multiplied by the constant 1 and the complication influence similarity), thereby realizing the logic in the foregoing analysis. When the treatment area similarity is larger, the tolerance sensitivity similarity occupies a larger proportion, and vice versa. When the treatment area similarity is smaller, the complication influence similarity occupies a larger proportion. Finally, the normalized value of the weighted result (the sum of the two products) is used as the linkage feature performance degree between the two sample data. The linkage feature performance degree adjusts the relative importance of the tolerance sensitivity similarity and the treatment area similarity by using the similar characteristics of the treatment area, and comprehensively considers multiple indexes. Therefore, the value can more comprehensively reflect the overall similarity between the sample data in the piezoelectric shock wave treatment process. The larger the value is, the higher the similarity is. The normalization is a technology known to those skilled in the art. The normalization function can be linear normalization or standard normalization. The specific normalization method is not limited herein.

[0112] The treatment control parameter determination module 104 is configured to train a neural network based on the linkage feature performance degree between the sample data and the treatment control parameters, so as to determine the treatment control parameters of the to-be-tested sample.

[0113] In the foregoing modules, the various individualized difference characteristics of the patients represented by the sample data are fused, so that the linkage feature performance degree between any two sample data is analyzed and obtained, which can comprehensively and accurately reflect the overall similarity between the sample data. Therefore, in this module, the neural network can be trained based on the linkage feature performance degree between the sample data and the treatment control parameters of the sample data, so as to determine the treatment control parameters of the to-be-tested sample by means of the strong learning ability and prediction ability of the neural network.

[0114] Preferably, in an embodiment of the present application, the neural network is trained based on the linkage feature performance degree between the sample data and the treatment control parameters, so as to determine the treatment control parameters of the to-be-tested sample, including:

[0115] The linkage feature performance degrees between all the historical samples are used to perform cluster analysis on all the historical samples, to obtain cluster clusters: the K-means clustering algorithm and a preset optimal K value are used to perform cluster analysis on all the historical sample data, to obtain all the cluster clusters. At this time, the historical samples in each cluster cluster have high similar characteristics. The distance measurement is the value after negative correlation mapping of the linkage feature performance degree between the historical sample data. The negative correlation mapping here can adopt the formula or wherein, denotes an exponential function with the natural constant e as the base, and x denotes an independent variable.

[0116] Then in each cluster, the average of the linkage feature performance degree between each historical sample and other historical samples is taken as the treatment performance feature value of each historical sample, and the treatment performance feature value of each historical sample is taken as the input of the neural network, the average of each historical sample under each treatment control parameter is calculated as the control parameter factor, and the control parameter factor of each historical sample under all kinds of treatment control parameters is taken as the output, so as to train the neural network, and obtain the trained neural network. At this time, each cluster obtained based on the historical sample clustering analysis corresponds to a trained neural network.

[0117] Finally, in all linkage feature performance degrees corresponding to the to-be-tested sample, the to-be-tested sample is divided into the cluster to which the historical sample with the maximum linkage feature performance degree belongs, and the average of the linkage feature performance degree between the to-be-tested sample and all historical samples in the cluster to which the to-be-tested sample belongs is taken as the input of the trained neural network corresponding to the cluster to which the to-be-tested sample belongs, so as to obtain the specific value of each treatment control parameter corresponding to the to-be-tested sample.

[0118] It should be noted that the preset optimal K value can be determined based on the silhouette coefficient method, and the silhouette coefficient method and the K-means clustering algorithm are known technologies, and the specific process is not described here. The training process of the neural network is a known technology, and the specific process is not described here. The neural network can specifically use CNN or RNN, and is not limited or described here.

[0119] In summary, the data acquisition module can systematically collect patient sample data of the same disease type, and the sample data is divided into historical samples and to-be-tested samples, covering multi-dimensional information such as treatment feedback text, treatment area characteristic parameters, treatment control parameters, and pain test data. Since patients have individual differences, the pain tolerance and feedback after treatment of each person are different, and therefore, in the sample data analysis module, the tolerance sensitivity of each sample data is accurately obtained by analyzing the change trend and similar characteristics of the numerical value of the pain test data in the sample data; meanwhile, the complication influence degree is accurately determined by comparing the similarity of the treatment feedback text of the sample data; the module can accurately grasp the individual physiological and pathological characteristics of the patient, provides a key basis for personalized treatment, and helps to improve the pertinence and effectiveness of treatment. Further, in the linkage feature analysis module, the similarity of the treatment area characteristic parameters, the tolerance sensitivity, and the complication influence degree between the sample data is comprehensively considered to determine the linkage feature performance degree between any two sample data. Thus, by mining the internal relationship and similar rules between different sample data, it is helpful to more reasonably refer to the historical samples in the subsequent process, and to provide a more scientific reference for setting the treatment control parameters of the to-be-tested sample. Therefore, in the treatment control parameter determination module, the linkage feature performance degree between the sample data and the treatment control parameter training neural network are based on, and the strong learning and prediction ability of the neural network can accurately determine the treatment control parameters of the to-be-tested sample according to the individual characteristics of the patient and the linkage relationship between the sample data. Therefore, the embodiment of the present application can fully consider the individual differences and similar characteristics between the sample data, and can accurately set the treatment control parameters, effectively avoiding the occurrence of excessive or insufficient treatment.

[0120] Referring to Figure 4 Fig. 1 shows a system structure schematic diagram of a depth controllable piezoelectric shock wave treatment system based on automatic focusing provided by an embodiment of the present application, which comprises a processor 400, a memory 401, a bus 402 and a communication interface 403, the processor 400, the communication interface 403 and the memory 401 are connected through the bus 402; wherein the memory 401 can contain a high-speed random access memory, the bus 402 can be an ISA bus, a PCI bus or an EISA bus, etc., the processor 400 can be an integrated circuit chip with signal processing capability; the memory 401 stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to realize the steps of each module in the depth controllable piezoelectric shock wave treatment system based on automatic focusing.

[0121] It is to be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim. In a claim, the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. It is further noted that claims may

[0122] The various embodiments in the specification are described in progressive order, and each embodiment can refer to the same or similar parts. Each embodiment is distinguished from other embodiments by the points that are emphasized in each embodiment.

[0123] The above description is merely illustrative of the application and is not to be taken in a limiting sense. It is contemplated that departures from the specific design choices disclosed can still come within the scope of the application.

Claims

1. A depth controllable piezoelectric shock wave therapy system based on automatic focusing, characterized in that, The system comprises: a data acquisition module, configured to acquire sample data of patients with the same disease type, the sample data comprising historical sample data and a to-be-tested sample, and each sample data comprising treatment feedback texts of a patient at each time of treatment, treatment area characteristic parameters, treatment control parameters, and pain test data of the patient; a sample data analysis module, configured to analyze a change trend and a numerical similarity feature of the pain test data in the sample data, to obtain a tolerance sensitivity of each sample data, and to determine a complication influence degree of each sample data by comparing similarities between the treatment feedback texts in each sample data; a linkage feature analysis module, configured to comprehensively determine a linkage feature performance degree between any two sample data based on similarities of the treatment area characteristic parameters, similarities of the tolerance sensitivities, and similarities of the complication influence degrees between the sample data; a treatment control parameter determination module, configured to train a neural network based on the linkage feature performance degrees between the sample data and the treatment control parameters, so as to determine treatment control parameters of the to-be-tested sample.

2. The depth controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 1, wherein, The method for obtaining the tolerance sensitivity comprises: The pain test data comprises subjective pain scores under each test energy and electromyographic signal data, and the test energy gradually increases; In each sample data, a subjective pain sensitivity of each sample data is determined according to a change trend of the subjective pain scores; similarities and numerical features of the electromyographic signals between the sample data are analyzed to determine an objective pain sensitivity of each sample data; a value obtained by multiplying the subjective pain sensitivity and the objective pain sensitivity of each sample data after normalization is taken as the tolerance sensitivity of each sample data.

3. The depth controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 2, wherein, The method for obtaining the subjective pain sensitivity comprises: In each sample data, the subjective pain scores are sorted according to an increasing order of the test energy to obtain a sorting sequence; in the sorting sequence, a ratio of each subjective pain score to a previous subjective pain score is calculated as a pain increase perception degree of each subjective pain score; in each sample data, the subjective pain scores are fused by using the pain increase perception degrees of the subjective pain scores, and a value obtained by normalizing the fused result is taken as the subjective pain sensitivity of each sample data.

4. The depth controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 2, wherein, The method for obtaining the objective pain sensitivity comprises: In the pain test data of each sample data, electromyographic signal data under each test energy and an absolute area of a horizontal axis are taken as electromyographic energy representation values; for any two sample data, a difference feature of the electromyographic energy representation values of the two sample data under the same test energy is analyzed, and an objective stimulation performance difference degree between the two sample data is obtained in combination with a test energy numerical feature; all the sample data are analyzed by clustering based on the objective stimulation performance difference degrees between the sample data, so as to obtain all clustering clusters; in each clustering cluster, a mean value of peak values of the electromyographic signals of all the sample data under all the test energy is normalized to obtain an objective pain sensitivity of each sample data in each clustering cluster.

5. The depth controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 4, wherein, The method for obtaining the objective stimulation performance difference degree comprises: For any two sample data, the absolute value of the difference between the myoelectricity energy characteristic values of the two sample data under the same test energy is taken as the myoelectricity difference characteristic value of the two sample data under the same test energy; The ratio of each test energy to the sum of all test energies is taken as the stimulation intensity weight of each test energy; For any two sample data, the myoelectricity difference characteristic value of the two sample data under the same test energy is weighted and fused by using the stimulation intensity weight of the test energy, and the normalized value of the obtained weighted result is taken as the objective stimulation performance difference degree of the two sample data.

6. The depth controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 1, wherein, The method for obtaining the complication influence degree comprises: Complication keyword extraction is performed on the treatment feedback text of each sample data at each treatment, to obtain a complication keyword set at each treatment; In each sample data, the Jaccard correlation coefficient between the complication keyword set at the last treatment and the complication keyword set at each previous treatment is calculated as a complication similarity factor; The sum of all complication similarity factors corresponding to each sample data is negatively correlated and mapped, and the value after the mapping is taken as the complication influence degree of each sample data.

7. The depth-controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 1, wherein, The method for obtaining the linkage feature performance degree comprises: Similar features of the treatment area feature parameters between any two sample data are analyzed to obtain a treatment area similarity between the two sample data; The absolute value of the difference between the tolerance sensitivity of the two sample data is negatively correlated and mapped and normalized, and the value after the mapping and normalization is taken as the tolerance sensitivity similarity between the two sample data; The absolute value of the difference between the complication influence degrees of the two sample data is negatively correlated and mapped and normalized, and the value after the mapping and normalization is taken as the complication influence similarity between the two sample data; The tolerance sensitivity similarity between the two sample data is weighted by using the treatment area similarity between the two sample data, the complication influence similarity between the two sample data is weighted by using a constant 1 minus the difference between the treatment area similarities, and the normalized value of the weighted result is taken as the linkage feature performance degree between the two sample data.

8. The depth-controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 7, wherein, The method for obtaining the treatment area similarity comprises: The treatment area feature parameters at least include the skin-to-target distance and the subcutaneous fat layer thickness, and the treatment area feature parameters of each sample data at each treatment are combined to form a feature parameter vector; For any two sample data, the Euclidean distance between any two feature parameter vectors between the two sample data is calculated, negatively correlated and mapped and normalized to obtain a treatment area similarity factor between the two sample data, and the mean value of all treatment area similarity factors between the two sample data is taken as the treatment area similarity.

9. The depth-controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 1, wherein, The method for training a neural network based on the linkage feature performance degree between sample data and the treatment control parameter, so as to determine the treatment control parameter of a to-be-tested sample, comprises: Based on the linkage feature performance degree between all historical samples, clustering analysis is performed on all historical samples to obtain clustering clusters; In each clustering cluster, the mean value of the linkage feature performance degree between each historical sample and other historical samples is taken as the treatment performance characteristic value of each historical sample; The method for training a neural network based on the linkage feature performance degree between sample data and the treatment control parameter, so as to determine the treatment control parameter of a to-be-tested sample, comprises: Based on the linkage feature performance degree between all historical samples, clustering analysis is performed on all historical samples to obtain clustering clusters; In each clustering cluster, the mean value of the linkage feature performance degree between each historical sample and other historical samples is taken as the treatment performance characteristic value of each historical sample; In each cluster, the treatment performance characteristic value of each historical sample is taken as an input of the neural network, the mean value of each historical sample under each treatment control parameter is calculated as a control parameter factor, and the control parameter factors of each historical sample under all kinds of treatment control parameters are taken as outputs, so that the neural network is trained to obtain a trained neural network; The to-be-tested sample is divided into the cluster to which the historical sample corresponding to the maximum linkage feature performance degree belongs, and the mean value of the linkage feature performance degree between the to-be-tested sample and all historical samples in the cluster is taken as an input of the trained neural network corresponding to the cluster, so that the value of each treatment control parameter corresponding to the to-be-tested sample is obtained.

10. The depth-controllable piezoelectric shock wave therapy system based on automatic focusing according to claim 9, wherein, The clustering analysis of all historical samples based on the linkage feature performance degree between the historical samples is performed to obtain the cluster, and the clustering analysis comprises: In all historical samples, the clustering analysis of all historical sample data is performed based on a K-means clustering algorithm and a preset optimal K value to obtain all clusters, wherein the distance measurement is a value obtained by performing a negative correlation mapping on the linkage feature performance degree between the historical sample data.

Images