Temporomandibular joint vibration mechanical wave acquisition and recognition system

Abstract

The invention relates to a temporomandibular joint vibration mechanical wave acquisition and recognition system, which belongs to the technical field of medical instruments and comprises a sensor assembly formed by sequentially stacking a transmission part, a piezoelectric film and a fixing part, and the sensor assembly is attached to a temporomandibular joint part of a human body and generates a simulated piezoelectric signal based on an acquired vibration mechanical wave signal. The analog piezoelectric signals are transmitted to the signal transmission assembly through the connection assembly, are amplified and converted into digital piezoelectric signals, and then are input into the signal identification assembly; and outputting a result of identifying whether the temporomandibular joint vibration mechanical wave is abnormal or not based on the input digital piezoelectric signal. The problems that in the prior art, temporomandibular joint examination and collection signals are poor in reliability, and accuracy and objectivity are limited due to dependence on manual recognition are solved.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a temporomandibular joint vibration mechanical wave acquisition and recognition system. Background Technology

[0002] Traditional methods for examining the temporomandibular joint (TMJ) rely heavily on the doctor's clinical experience and the patient's subjective feelings, which limits their accuracy and objectivity. Utility model CN ​​212307887U discloses an existing technology for a temporomandibular joint movement sound acquisition device, employing a structure similar to a traditional stethoscope. While simple and easy to implement, this method relies on acquiring sound waves generated by TMJ movement and then having a doctor manually interpret them. However, the weak sound signals generated by TMJ movement are prone to distortion, and the inherent uncertainty of manual diagnosis makes it difficult to guarantee the accuracy of sound wave acquisition combined with manual judgment. Another existing patent, 201880092230.1, discloses a joint analysis probe that uses a microphone-like structure to isolate external environmental noise and acquire sound wave signals generated by the TMJ. This also suffers from low signal acquisition accuracy and has a complex structure, still relying heavily on manual intervention, making accurate signal recognition equally difficult to guarantee.

[0003] In the study of temporomandibular joint disorders, by collecting the mechanical wave signals generated by the vibration of the temporomandibular joint during movement, and identifying the disorder based on the collected signals, we can gain a more comprehensive understanding of the joint's vibration characteristics, stress distribution, and pathological changes, providing important evidence for the early detection, diagnosis, and treatment of the disease. Summary of the Invention

[0004] Based on the above analysis, the present invention aims to provide a temporomandibular joint vibration mechanical wave acquisition and identification system to solve the problems of poor reliability of existing temporomandibular joint examination acquisition signals and the limitations of accuracy and objectivity due to reliance on manual identification.

[0005] On one hand, embodiments of the present invention provide a temporomandibular joint vibration mechanical wave acquisition and recognition system, the system comprising a sensor assembly, a connection assembly, a signal processing assembly, and a signal recognition assembly, wherein,

[0006] The sensor assembly is used to acquire analog piezoelectric signals of the mechanical vibration waves of the temporomandibular joint;

[0007] The analog piezoelectric signal is transmitted to the signal processing component through the connection component. After signal amplification and analog-to-digital conversion, a digital piezoelectric signal is output to the signal recognition component.

[0008] The signal recognition component preprocesses the input digital piezoelectric signal and then uses a trained random forest model to output the recognition result of whether the temporomandibular joint vibration mechanical wave is abnormal based on the preprocessed digital piezoelectric signal.

[0009] The beneficial effects of the above technical solution are as follows: it can effectively collect the mechanical vibration waves of the temporomandibular joint and automatically identify whether the mechanical vibration waves of the temporomandibular joint are abnormal.

[0010] Based on further improvements to the above system, the sensor assembly includes a transmission component, a piezoelectric film, and a fixing component stacked together, wherein...

[0011] The transmission component is used to transmit the mechanical vibration waves of the temporomandibular joint to the piezoelectric film.

[0012] The piezoelectric film is used to detect the vibrational mechanical waves transmitted by the transmission component and trigger the generation of a simulated piezoelectric signal;

[0013] The piezoelectric film and the transmission component are stacked sequentially on the fixed component.

[0014] The beneficial effect of the above-mentioned further improvement scheme is that the sensor component structure can transmit the mechanical waves of temporomandibular joint vibration to the piezoelectric film to generate an analog electrical signal for abnormal detection.

[0015] Based on further improvements to the above system, the transmission component specifically comprises:

[0016] Made of skin-friendly insulating material with good mechanical wave conduction capability;

[0017] The upper and lower surfaces are parallel, with the upper surface in contact with the human body and the lower surface in contact with the piezoelectric film.

[0018] The lower surface has the same shape as the piezoelectric film, but its cross-sectional area is slightly smaller than that of the piezoelectric film.

[0019] The edges are bonded and fixed to the piezoelectric film by center alignment with the piezoelectric film.

[0020] The beneficial effect of the above-mentioned further improvement scheme is that the structure and arrangement of the transmission component are conducive to better acquisition of the mechanical vibration waves of the temporomandibular joint and transmission to the piezoelectric film.

[0021] Based on further improvements to the above system, the sensor assembly also includes metal contacts for conforming to the skin and reducing signal interference.

[0022] The beneficial effects of the above-mentioned further improvement scheme are that the structure and arrangement of the metal contact not only facilitate the sensor assembly to fit better with the human body, but also help to reduce signal interference.

[0023] Based on the further improvement of the above system, the fixing component is specifically: the upper and lower surfaces are parallel, wherein the groove or protrusion is provided on one side of the upper surface, the cross-sectional shape of the groove or protrusion is the same as that of the piezoelectric film, the area is smaller than that of the piezoelectric film, the piezoelectric film covers the groove or protrusion, and the edge is bonded to the fixing component; the metal contact is fixedly provided on the other side of the upper surface.

[0024] The beneficial effects of the above-mentioned further improvement scheme are: the structure and arrangement of the fixed component ensure the overall structural stability of the sensor assembly, which is conducive to the piezoelectric film generating simulated piezoelectric signals better based on vibration mechanical waves.

[0025] Based on further improvements to the above system, the upper surface of the metal contact is at the same height as the upper surface of the transmission component, and is configured with a smooth surface and edge structure; the metal contact is also connected to the signal processing component through the connecting component.

[0026] The beneficial effects of the above-mentioned further improvement scheme are that the structure and arrangement of the metal contact not only facilitate the sensor assembly to fit better with the human body, but also help to reduce signal interference.

[0027] Based on further improvements to the above system, a trained random forest model is obtained through the following method, which specifically includes:

[0028] The raw signal samples used for model training are preprocessed and labeled to construct training and test sets;

[0029] The number of decision trees and the minimum number of leaf samples are used as model hyperparameters. An initial set of hyperparameter values ​​is set, and a set of hyperparameter values ​​is selected from each initial set to generate multiple combinations of hyperparameter values. Multiple random forest models to be trained are generated based on each combination of hyperparameter values.

[0030] The training set is used to train the random forest model to be trained, and then each trained random forest model is generated.

[0031] The fitness values ​​of each trained random forest model are calculated based on the training and test sets, and the random forest model with the best fitness value is selected as the trained random forest model.

[0032] The beneficial effect of the above-mentioned further improvement scheme is that the method can obtain a random forest model with good ability to identify whether the mechanical waves of temporomandibular joint vibration are abnormal.

[0033] Based on further improvements to the above system, preprocessing is performed using the following method, which specifically includes:

[0034] The original signal sample is denoised to obtain the denoised signal.

[0035] Extract feature parameters from the noise-reduced signal;

[0036] Normalized feature parameters are generated by normalizing the feature parameters respectively.

[0037] The beneficial effects of the above-mentioned further improvement scheme are: the feature parameters extracted by the preprocessing method help improve the ability of the random forest model to identify whether the mechanical vibration wave of the temporomandibular joint is abnormal; the noise reduction process improves the accuracy and effectiveness of the signal; and the normalization of the extracted feature parameters can reduce the data processing difficulty and complexity of the random forest model training without affecting the model training effect.

[0038] Based on the further improvement of the above system, the step of calculating the fitness value of each trained random forest model based on the training set and the test set, and selecting the random forest model with the best fitness value as the trained random forest model specifically includes:

[0039] The training set and the test set are respectively input into the trained random forest model, and the recognition results are output respectively.

[0040] Obtain the error between the recognition result and the labeled value of each sample, and obtain the error of the training set based on the error of all samples in the training set, and obtain the error of the test set based on the error of all samples in the test set; use the sum of the errors of the training set and the test set as the recognition error value of the random forest model.

[0041] The ratio of the recognition error value of each trained random forest model to the total number of samples in the input training and test sets is calculated as the fitness value of each trained random forest model.

[0042] The trained random forest model with the smallest fitness value is used as the trained random forest model.

[0043] The beneficial effect of the above-mentioned further improvement scheme is that the fitness value calculation helps to select the random forest model with the best recognition ability from all trained random forest models as the trained random forest model, thereby obtaining the best recognition effect for the mechanical waves of temporomandibular joint vibration.

[0044] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0045] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0046] Figure 1 This is a schematic diagram of the system structure according to an embodiment of the present invention.

[0047] Figure 2 This diagram illustrates the connection relationships between the components of the system and an example of how it is worn by a human body, according to an embodiment of the present invention.

[0048] Figure 3 This is a schematic diagram of the sensor assembly structure according to an embodiment of the present invention.

[0049] Figure label:

[0050] A: Sensor assembly; B: Connection assembly; C: Processing and transmission assembly; D: Signal recognition assembly; 1-Transmission component; 2-Piezoelectric film; 3-Fixing component; 4-Metal contact; 5-Grounding wire; 6-Structural component; 7-Signal line. Detailed Implementation

[0051] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0052] A specific embodiment of the present invention discloses a temporomandibular joint vibration mechanical wave acquisition and identification system, such as... Figure 1 As shown.

[0053] The system includes a sensor assembly, a connection assembly, a signal processing assembly, and a signal recognition assembly, wherein,

[0054] The sensor assembly is used to acquire analog piezoelectric signals of the mechanical vibration waves of the temporomandibular joint;

[0055] The analog piezoelectric signal is transmitted to the signal processing component through the connection component. After signal amplification and analog-to-digital conversion, a digital piezoelectric signal is output to the signal recognition component.

[0056] The signal recognition component preprocesses the input digital piezoelectric signal and then uses a trained random forest model to output the recognition result of whether the temporomandibular joint vibration mechanical wave is abnormal based on the preprocessed digital piezoelectric signal.

[0057] Specifically, this embodiment discloses a temporomandibular joint mechanical vibration wave acquisition and recognition system. The system acquires mechanical vibration waves by fitting the sensor assembly to the temporomandibular joint of the human body. The sensor assembly processes the acquired temporomandibular joint mechanical vibration waves into simulated piezoelectric signals. The sensor assembly is then connected to the connecting assembly, and the connecting assembly is connected to the signal transmission assembly, forming a connection structure. This structure not only transmits the simulated piezoelectric signals to the signal transmission assembly through the connecting assembly, but also allows the sensor assembly to be worn well on the human body and fit the temporomandibular joint.

[0058] Figure 2 The document illustrates several implementation connections and human wearing methods for the temporomandibular joint mechanical vibration wave acquisition and recognition system of this embodiment. In one embodiment, the sensor assembly fits snugly against the temporomandibular joint area of ​​the human body. The transmission assembly can be placed on the top of the head or behind the neck and connected to the sensor assembly via the connecting assembly to form a head-mounted or neck-hook-style structure that can be fixed to the human body. Alternatively, the transmission assembly can be placed without contact with the human body and form a separate structure with the sensor assembly via the connecting assembly. The signal recognition assembly is connected to the transmission assembly but does not contact the human body. In a preferred embodiment, the signal recognition assembly is connected to the transmission assembly wirelessly. The connecting assembly includes a structural component, a signal line, and a grounding wire, wherein the signal line and grounding wire are disposed within the cavity of the structural component. It should be noted that... Figure 2 The examples provided are not intended to limit the system implementation methods and scope disclosed in this embodiment.

[0059] like Figure 3 As shown, the sensor assembly includes a transmission component, a piezoelectric film, and a fixing component stacked together, wherein...

[0060] The transmission component is used to transmit the mechanical vibration waves of the temporomandibular joint to the piezoelectric film.

[0061] The piezoelectric film is used to detect the vibrational mechanical waves transmitted by the transmission component and trigger the generation of a simulated piezoelectric signal;

[0062] The piezoelectric film and the transmission component are stacked sequentially on the fixed component.

[0063] Specifically, the sensor assembly collects mechanical vibration waves generated by temporomandibular joint movement, causing a piezoelectric film to vibrate. The piezoelectric film then generates a simulated piezoelectric signal by precisely sensing the vibration, thus achieving signal acquisition of the mechanical vibration waves. To achieve this technical objective, the transmission component specifically comprises:

[0064] Made of skin-friendly insulating material with good mechanical wave conduction capability;

[0065] The upper and lower surfaces are parallel, with the upper surface in contact with the human body and the lower surface in contact with the piezoelectric film.

[0066] The lower surface has the same shape as the piezoelectric film, but its cross-sectional area is slightly smaller than that of the piezoelectric film.

[0067] The edges are bonded and fixed to the piezoelectric film by center alignment with the piezoelectric film.

[0068] The fixing component specifically comprises: upper and lower surfaces parallel to each other, wherein the groove or protrusion is provided on one side of the upper surface, the cross-sectional shape of the groove or protrusion is the same as that of the piezoelectric film, the area is smaller than that of the piezoelectric film, the piezoelectric film covers the groove or protrusion, and the edge is bonded to the fixing component.

[0069] Specifically, the upper and lower surfaces of the transmission component are parallel. The upper surface, which contacts the human body, is made of a skin-friendly insulating material to reduce discomfort and prevent noise signals from being transmitted from the human body's bioelectricity to the piezoelectric film. The lower surface contacts the piezoelectric film, has the same shape as the film, and has a slightly smaller cross-sectional area. It is aligned with the piezoelectric film at the center, and its edges are bonded to the upper surface of the film. The upper and lower surfaces of the fixing component are parallel, preferably a cube made of a regular and rigid material. The upper and lower surfaces of the fixing component have the same area. One side of the upper surface has a groove or protrusion with the same shape as the piezoelectric film, and the area of ​​the groove or protrusion is smaller than that of the film. The lower surface of the piezoelectric film covers the groove or protrusion, and its lower surface edge is bonded to the upper surface of the fixing component. This ensures a stable structure with the piezoelectric film and facilitates the formation of a good cavity effect, which is beneficial for the piezoelectric film to better generate simulated piezoelectric signals based on vibrational mechanical waves.

[0070] Figure 3 As shown, the sensor assembly further includes a metal contact for conforming to the skin and reducing signal interference; furthermore, the upper surface of the metal contact is at the same height as the upper surface of the transmission component and is configured with a smooth surface and edges; the metal contact is also connected to the signal processing component via the connecting component.

[0071] Specifically, the metal contact is fixed to the other side of the upper surface of the fixing component and does not contact the piezoelectric film. By setting the metal contact, the sensor assembly forms a structure with parallel upper and lower surfaces and similar areas, which allows for a more stable fit to the human body during application.

[0072] The metal contacts are also connected to the signal transmission component. For example... Figure 1As shown, the signal transmission component includes a signal amplification module, an analog-to-digital conversion module, and a signal transmission module. The signal amplification module receives the analog piezoelectric signal output by the sensor component through the connection component. To reduce signal interference, the signal amplification module is also connected to the metal contact. The metal contact conforms to the human body to form a common-ground effect, thereby reducing noise interference input to the signal amplification module. After signal amplification, the analog piezoelectric signal is output to the analog-to-digital conversion module, which converts the analog piezoelectric signal into a digital piezoelectric signal, which is then output to the signal recognition component through the signal transmission module.

[0073] The signal recognition component first preprocesses the input digital piezoelectric signal using the following method, which specifically includes:

[0074] The original signal sample is denoised to obtain the denoised signal.

[0075] Extract feature parameters from the noise-reduced signal;

[0076] Normalized feature parameters are generated by normalizing the feature parameters respectively.

[0077] Specifically, the signal recognition component first performs noise reduction processing on the input digital piezoelectric signal. Preferably, the noise reduction method adopts the mean filtering method, that is, summing the digital piezoelectric signals within a certain time window, and then dividing by the time window length to obtain the mean value, so as to reduce the signal processing complexity and eliminate noise interference. Optionally, the noise-reduced signal can also be obtained by median filtering, low-pass filtering, or weighted mean filtering, but these are not preferred for the technical solution of this embodiment.

[0078] Next, feature parameters are extracted from the denoised signal. In this embodiment, the root mean square value and the sum of waveform lengths of the denoised signal are extracted as time-domain feature parameters, and the median frequency and the mean power frequency distribution are extracted as frequency-domain feature parameters to determine whether the temporomandibular joint vibration mechanical wave is abnormal. Extensive experimental verification shows that selecting the root mean square value, the sum of waveform lengths, the median frequency, and the mean power frequency distribution of the denoised signal as feature parameters to determine whether the temporomandibular joint vibration mechanical wave is abnormal yields very satisfactory results.

[0079] Specifically, the root mean square value refers to the square root of the average signal amplitude within a certain time window, and is expressed by the following formula:

[0080] In the formula, RMS is the root mean square of the signal amplitude, x i The signal is denoised, and N is the set time window.

[0081] The waveform length summation refers to the summation of signal waveform lengths within a certain time window N, specifically expressed by the following formula:

[0082] The larger the sum of the waveform lengths, the more drastic the change in the signal waveform.

[0083] The median frequency is obtained by analyzing the power spectral density of the signal. Power spectral density is a measurement index used to describe the distribution of signal power at different frequencies, representing the average power per unit frequency band (e.g., per Hertz). It is calculated by performing a Fourier transform on the signal, specifically as follows:

[0084] PSD(f)=|X(f)| 2 In the formula, X(f) refers to the amplitude value obtained after the noise-reduced signal undergoes Fourier transform, and f represents the frequency;

[0085] The median frequency is specifically expressed by the following formula:

[0086] In the formula, MF represents the median frequency, and df represents the derivative of the frequency;

[0087] The mean value of the signal power frequency distribution is specifically expressed by the following formula:

[0088] In the formula, MPF is the mean power frequency distribution.

[0089] Furthermore, the feature parameters are normalized, preferably using the minimax normalization method. The feature parameters are mapped to the [0, 1] interval using the following formula, which aims to reduce the complexity of data processing. The formula is specifically expressed as follows:

[0090] In the formula, x is the characteristic parameter, x new The normalized feature parameters, x min x is the minimum value among the feature parameter categories. max The maximum value among the aforementioned feature parameter categories. Optionally, Z-score normalization and Sigmoid normalization methods can also be used for data normalization in this embodiment, but they are not preferred after experimental verification.

[0091] At this point, the signal recognition component has completed the preprocessing of the digital piezoelectric signal. The signal recognition component also includes a trained random forest model. Inputting the preprocessed digital piezoelectric signal into the trained random forest model will output an identification result indicating whether the corresponding temporomandibular joint vibration mechanical wave is abnormal. The trained random forest model is obtained through the following method, specifically including:

[0092] The raw signal samples used for model training are preprocessed and labeled to construct training and test sets;

[0093] The number of decision trees and the minimum number of leaf samples are used as model hyperparameters. An initial set of hyperparameter values ​​is set, and a set of hyperparameter values ​​is selected from each initial set to generate multiple combinations of hyperparameter values. Multiple random forest models to be trained are generated based on each combination of hyperparameter values.

[0094] The training set is used to train the random forest model to be trained, and then each trained random forest model is generated.

[0095] The fitness values ​​of each trained random forest model are calculated based on the training and test sets, and the random forest model with the best fitness value is selected as the trained random forest model.

[0096] The structure and performance of a random forest model are closely related to the selected hyperparameters. In this embodiment, the number of decision trees and the minimum number of leaf samples are selected as the hyperparameters of the random forest model. In order to obtain the optimal hyperparameter values ​​of the random forest model, i.e. the optimal number of decision trees and the minimum number of leaf samples, this embodiment adopts an exhaustive search method. Specifically, it means: based on experience, several candidate values ​​for decision trees and several candidate values ​​for the minimum number of leaf samples are preset, and then the candidate values ​​for decision trees and the minimum number of leaf samples are combined to verify the performance of the random forest model constructed. That is, the number of random forest models needs to be constructed as many times as there are candidate values ​​for decision trees and the minimum number of leaf samples. The random forest model with the best performance is selected as the trained random forest model.

[0097] First, the signal recognition component performs the same noise reduction, feature parameter extraction, and normalization preprocessing on the original signal samples used for model training as on the input digital piezoelectric signal. Then, the original signal samples are labeled. Specifically, each original signal sample is labeled with a tag indicating whether it is abnormal. For example, a normal signal sample is labeled as 0, and an abnormal signal sample is labeled as 1. The labeling method is consistent with the final recognition result format. To ensure the correctness of the labeling of the original signal samples, expert manual labeling is preferred. The labeled original signal samples are used to construct training and test sets. Specifically, the construction of training and test sets involves randomly sampling the original signal samples the same number of times as the number of decision trees. Each time, 70% of the original signal samples are sampled to create a training set, and the remaining 30% is used to create a corresponding test set, resulting in multiple training and test sets. The number of decision trees selected here is one of the alternative values ​​for decision trees. The purpose of randomly sampling 70% of the original signal samples with replacement to create the training set is to enhance the randomness of the samples in the constructed training set, which is beneficial to improving the robustness of the random forest model after training. For each decision tree candidate value and minimum leaf sample number combination, a corresponding training set and test set must be established, and the number of the training set and test set is the same as the number of the corresponding decision tree candidate values.

[0098] Next, the training set is used to train the random forest model to be trained, and each trained random forest model is generated by the following method, which specifically includes:

[0099] In the training set, feature parameters of 75% of the categories of each sample are randomly selected and input into the root node of a decision tree in the random forest model to be trained.

[0100] The sample entropy before splitting is calculated based on the sample feature parameters of the input training set, and a decision tree is established through the following steps, which specifically include:

[0101] S1: Sort the feature parameters of each category of each sample in descending order;

[0102] S2: Calculate the split value sequentially based on the comparison results of adjacent feature parameters in each queue;

[0103] S3: Split the current root node into two leaf nodes;

[0104] S4: Based on the comparison results between each split value and the corresponding category feature parameter of each sample, each sample is divided into two groups. The samples whose corresponding category feature parameter is less than the split value are divided into one group, and the remaining samples are divided into another group. The entropy of each feature parameter after grouping is calculated.

[0105] S5: Save the difference between the entropy before splitting and the entropy after grouping of each sample feature parameter as the gain, associate the corresponding category feature with the largest gain with the current root node, and use the splitting value corresponding to the feature parameter with the largest gain as the splitting value of the current root node.

[0106] S6: Based on the split value grouping result of the current root node, the two groups of samples are respectively assigned to a leaf node of the current root node, and the feature parameters corresponding to the feature category with the largest gain are removed from each sample;

[0107] For each leaf node sample, S1-S6 are executed iteratively. In each iteration, the feature parameters of the category features of the associated nodes are ignored until all category features are associated with a certain node, at which point the decision tree is completed.

[0108] Once all decision trees in the random forest model to be trained are built, the trained random forest model is obtained.

[0109] Specifically, for each decision tree candidate value and the minimum number of leaf samples candidate value to build a random forest model, the feature parameters of 75% of the corresponding training sets are randomly selected and input into the random forest model. In this embodiment, each original signal sample includes 4 types of feature parameters, and each training set randomly selects 3 types of feature parameters. The purpose is to limit the depth of the decision tree during the model training process while ensuring the difference in the feature parameters used by each decision tree, and to prevent overfitting.

[0110] In each random forest model, each training set randomly corresponds to the root node of a decision tree. Based on the characteristics of the random forest model, each node in the decision tree binary splits into two leaf nodes, and each node records and is associated with a feature parameter. In this embodiment, the entropy of each sample before and after the split is calculated using the following formula, and then the association between the node and the feature parameter is determined by calculating the gain of the entropy. The formula is specifically expressed as follows:

[0111] In the formula,

[0112] x i Represents characteristic parameters,

[0113] H(X) represents entropy.

[0114] p(x i ) represents the probability of a feature parameter value occurring, specifically x i The ratio of the number of times the value appears in the current node to the total number of samples in the current node.

[0115] The gain of entropy refers to the difference between the entropy after splitting the same sample feature parameters and the entropy before splitting.

[0116] Calculate and save the entropy of all sample parameters input to the root node;

[0117] Then, the sample parameters are sorted in descending order for each category of feature parameters. Specifically, in this embodiment, by sorting the feature parameters in descending order for each category, four kinds of sample descending order results are obtained. The purpose is to establish a sort for each category of feature parameters to facilitate pairwise verification and further enhance randomness.

[0118] The splitting value is calculated sequentially based on the comparison results of adjacent feature parameters in each queue, specifically including:

[0119] Each queue checks whether adjacent feature parameters are the same from the beginning. If they are the same, the comparison continues until adjacent feature parameters are different. The sum of 2 / 3 of the previous feature parameter and 1 / 3 of the next feature parameter is rounded to the nearest integer and saved as the split value.

[0120] Otherwise, the current feature parameter and the mean of adjacent feature parameters are rounded to the nearest integer and saved as the split value until each queue has completed the judgment of all adjacent feature parameters.

[0121] For example, if one of the characteristic parameters of the descending sequence is [7, 4, 4, 3, 3, 2, 1], then the calculated split value is [6, 4, 3, 2].

[0122] After processing, we can obtain the split values ​​based on the weighted average of the feature parameters of each pair of adjacent samples under the four types of sample descending order.

[0123] The next step is to find the optimal splitting value from among the various splitting values, specifically including:

[0124] Based on the comparison between each split value and the corresponding category feature parameter of each sample, the samples are divided into two groups. Specifically, only the category feature parameter of each sample with the calculated split value is compared with the corresponding split value. Samples with a category feature parameter smaller than the split value are divided into one group, and the remaining samples are divided into the other group. The entropy of each group is calculated. From the entropy calculation formula, it can be seen that when each sample is divided into two groups based on a feature, the probability p(x) of the feature parameter value of each sample before and after being divided into two groups is... iThe entropy after grouping will change, resulting in a difference between the entropy before and after grouping. Therefore, the entropy of each sample before and after grouping based on the split values ​​is calculated, and then each gain is calculated separately. The split value corresponding to the feature parameter with the largest gain is used as the split value of the current root node, thus completing the association between the current root node and a certain feature. At this time, the decision tree completes a split based on the split value of the current root node. The grouping results of each sample based on the split value of the current root node are assigned to a leaf node of the root node, and the feature parameter corresponding to the split value of the root node is removed from each sample. Then, the split value of each leaf node is recalculated, and the entropy before and after grouping of the corresponding samples is calculated and the maximum gain is found. The associated feature type of each leaf node is found, and then the split is performed downwards. This process is repeated until all feature types are associated with a certain node, and then the decision tree is built.

[0125] Once all decision trees are built, the random forest model training is complete, and the trained random forest models are obtained.

[0126] Generally, the performance of a random forest model can be evaluated by calculating its fitness value. In this embodiment, the random forest model with the smallest error after training is considered the optimal one.

[0127] Specifically, each of the trained random forest models has been trained on its respective input training set, learning and recording the features of the samples in the training set. Due to the randomness of the training set samples, as well as the randomness of the decision tree generation process and training of each trained random forest model, the performance of each trained random forest model is different. Therefore, the corresponding training set and test set are input again into each trained random forest model to output the recognition result. Specifically, in this embodiment, the random forest model outputs a qualitative recognition result on whether the input temporomandibular joint vibration mechanical wave signal is abnormal. The recognition result is then compared with the labeled value of the corresponding sample. If the recognition result is different from the labeled value of the corresponding sample, it is recorded as an error. The number of times the training set error and test set error occur for each trained random forest model are summarized and counted. The ratios are then divided by the number of samples in the training set and test set, and summed to obtain the error value for each trained random forest model. The random forest model with the smallest error value is the trained random forest model.

[0128] This embodiment discloses a temporomandibular joint (TMJ) vibration mechanical wave acquisition and identification system, comprising a sensor assembly consisting of a transmission component, a piezoelectric film, and a fixing component stacked sequentially. The sensor assembly is fitted to the TMJ region of the human body to generate a simulated piezoelectric signal based on the acquired vibration mechanical wave signal. This simulated piezoelectric signal is transmitted to the signal transmission component via a connecting component. After signal amplification and conversion into a digital piezoelectric signal, it is input to the signal identification component. The signal identification component uses a trained random forest model to output an identification result indicating whether the TMJ vibration mechanical wave is abnormal, based on the input digital piezoelectric signal. Compared to existing technologies, this embodiment acquires TMJ vibration mechanical wave signals through a unique sensor structure design. After processing, the abnormality result is identified by a trained random forest model. The random forest model is trained with optimized algorithms, ensuring reliable performance and eliminating the need for manual identification. This significantly improves the reliability of the acquired TMJ signal and solves the problem of limited accuracy and objectivity in manual identification.

[0129] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0130] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A temporomandibular joint vibration mechanical wave acquisition and identification system, characterized by, The system comprises a sensor component, a connection component, a signal processing component, and a signal identification component, wherein, the sensor component is used to collect analog piezoelectric signals of the vibration mechanical waves of the temporomandibular joint; the analog piezoelectric signals are transmitted to the signal processing component through the connection component, and after signal amplification and analog-digital conversion, digital piezoelectric signals are output to the signal identification component; the signal identification component pre-processes the input digital piezoelectric signals, and then outputs the identification result of whether the vibration mechanical waves of the temporomandibular joint are abnormal based on the pre-processed digital piezoelectric signals using the trained random forest model.

2. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 1, characterized in that, The sensor component comprises a transmission part, a piezoelectric film, and a fixing part arranged in layers, wherein, the transmission part is used to transmit the vibration mechanical waves of the temporomandibular joint to the piezoelectric film; the piezoelectric film is used to detect the vibration mechanical waves transmitted by the transmission part and trigger the generation of analog piezoelectric signals; the piezoelectric film and the transmission part are arranged in layers on the fixing part.

3. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 2, characterized in that, The transmission part is specifically: made of a skin-friendly insulating material with good mechanical wave conduction ability; the upper and lower surfaces are parallel, wherein the upper surface contacts the human body and the lower surface contacts the piezoelectric film; the lower surface has the same shape as the piezoelectric film, and the cross-sectional area is slightly smaller than the area of the piezoelectric film; the piezoelectric film is aligned with the center and is fixed by adhesion at the edge.

4. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 2, characterized in that, The sensor component further comprises a metal contact for adhering to the skin and reducing signal interference.

5. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 2, characterized in that, The fixing part is specifically:

6. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 2, characterized in that, the upper and lower surfaces are parallel, wherein one side of the upper surface is provided with the groove or protrusion, the cross-sectional shape of the groove or protrusion is the same as that of the piezoelectric film, and the area is smaller than that of the piezoelectric film; the piezoelectric film covers the groove or protrusion, and the edge is adhesively fixed to the fixing part; the other side of the upper surface is fixedly provided with the metal contact.

7. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 1, characterized in that, The upper surface of the metal contact has the same height as the upper surface of the transmission part, and is provided with a smooth surface and edge structure; the metal contact is also connected to the signal processing component through the connection component. The trained random forest model is obtained by the following method, which specifically comprises: pre-processing the original signal samples used for model training, and then constructing a training set and a test set after labeling; setting the decision tree number and the minimum leaf sample number as model hyperparameters, setting the initial value set of each hyperparameter, selecting the hyperparameter value from each initial value set to generate multiple hyperparameter value combinations, and generating multiple random forest models to be trained according to each hyperparameter value combination; training the random forest models to be trained respectively using the training set to generate each trained random forest model; 8. A system for collecting and identifying mechanical vibrations of the temporomandibular joint according to claim 1 or 7, characterized in that, calculating the fitness values of each trained random forest model based on the training set and the test set, and selecting the random forest model with the best fitness value as the trained random forest model. The pre-processing is performed by the following method, which specifically comprises: obtaining a denoised signal after noise reduction processing of the signal; extracting feature parameters from the denoised signal; generating normalized feature parameters after normalizing the feature parameters respectively.

9. A system for collecting and identifying mechanical vibrations in the temporomandibular joint according to claim 8, characterized in that, The training set and the test set are constructed, specifically, the original signal samples are randomly sampled the same number of times as the decision tree, and each time 70% of the original signal samples are sampled to establish a training set, and the remaining 30% is established as a corresponding test set, to obtain multiple training sets and test sets.

10. A system for collecting and identifying mechanical vibrations in a temporomandibular joint according to claim 9, characterized in that The fitness value of each trained random forest model is calculated based on the training set and the test set, and the random forest model with the best fitness value is taken as the trained random forest model, specifically including: The training set and the test set are input into the trained random forest model to output identification results, respectively; The error of the identification result corresponding to each sample and the labeled value of the sample is obtained, and the error of the training set is obtained based on the error of all samples in the training set, and the error of the test set is obtained based on the error of all samples in the test set; the sum of the errors of the training set and the test set is taken as the identification error value of the random forest model; The ratio of the identification error value of each trained random forest model to the total number of input training set and test set samples is calculated as the fitness value of each trained random forest model. The trained random forest model with the smallest fitness value is taken as the trained random forest model.

Images