Method for evaluating orthodontic treatment effect

Abstract

The invention relates to the technical field of medical information processing, in particular to an orthodontic treatment effect evaluation method. Comprising the following steps: acquiring and preprocessing multi-dimensional data in real time to obtain preprocessed multi-dimensional data; performing enhancement fusion processing on the preprocessed multi-dimensional data by using a similarity-based adaptive transformation enhancement fusion algorithm to obtain comprehensive multi-dimensional data; evaluating and analyzing the comprehensive multi-dimensional data by using an orthodontic treatment effect evaluation algorithm based on fuzzy reasoning to obtain an evaluation result; and evaluating the orthodontic treatment effect based on the evaluation result. The technical problems that multi-dimensional data processing is not accurate enough and evaluation accuracy is low in orthodontic treatment are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical information processing, and in particular to an orthodontic treatment effect evaluation method. BACKGROUND

[0002] Currently, the evaluation and adjustment of orthodontic treatment mostly rely on traditional imaging examination and clinical observation, usually including X-ray films, CT images, etc. However, the traditional evaluation method has obvious limitations. First, although imaging data can provide certain structural information, it cannot accurately reflect the feelings and feedback of patients during the treatment process, such as pain, comfort, and other subjective factors. In addition, there may be a certain delay in imaging data and clinical records in some cases, resulting in insufficient timely treatment adjustment.

[0003] Many hospitals and clinics have begun to try to use digital technology and artificial intelligence to optimize orthodontic treatment, but most of them focus on image processing, oral scanning, and other fields. Although these technologies can provide more detailed treatment plans and accurate digital models, their application still mainly relies on a single data source, such as 3D oral scanning data, digital X-ray images, etc. A single data source cannot fully reflect the treatment progress and potential treatment problems of patients, especially in terms of different patient treatment responses and individual differences. Relying on traditional treatment adjustment methods may lead to unstable treatment effects.

[0004] However, the above existing technology still has the following technical problems: insufficient accuracy in multi-dimensional data processing and low evaluation accuracy in orthodontic treatment. SUMMARY

[0005] The present application provides an orthodontic treatment effect evaluation method to solve the technical problems of insufficient accuracy in multi-dimensional data processing and low evaluation accuracy in orthodontic treatment.

[0006] The orthodontic treatment effect evaluation method of the present application specifically includes the following technical solutions: An orthodontic treatment effect evaluation method includes the following steps: S1. Real-time acquisition and preprocessing of multi-dimensional data to obtain preprocessed multi-dimensional data; using a similarity-based adaptive transformation enhancement fusion algorithm to enhance and fuse the preprocessed multi-dimensional data to obtain comprehensive multi-dimensional data; S2. Using a fuzzy inference-based orthodontic treatment effect evaluation algorithm to evaluate and analyze the comprehensive multi-dimensional data to obtain an evaluation result; based on the evaluation result, evaluating the orthodontic treatment effect.

[0007] Preferably, the S1 specifically includes: The similarity-based adaptive transformation enhancement fusion algorithm utilizes the similarity between features to describe the correlation of the preprocessed multi-dimensional data, and then performs enhancement fusion processing on the preprocessed multi-dimensional data through multi-scale adaptive transformation to obtain comprehensive multi-dimensional data.

[0008] Preferably, the S1 specifically includes: In the implementation process of the similarity-based adaptive transformation enhancement fusion algorithm, feature extraction is performed on the preprocessed multi-dimensional data to obtain initial feature data.

[0009] Preferably, the S1 specifically includes: In the implementation process of the similarity-based adaptive transformation enhancement fusion algorithm, based on the initial feature data, a similarity-based distance-weighted adjacency matrix and a space-time decay term are combined to construct an adjacency matrix representing the similarity between the initial feature data of each feature.

[0010] Preferably, the S1 specifically includes: In the implementation process of the similarity-based adaptive transformation enhancement fusion algorithm, based on the similarity between the initial feature data, nonlinear mapping is performed on each initial feature data to obtain nonlinearly mapped feature data.

[0011] Preferably, the S1 specifically includes: In the implementation process of the similarity-based adaptive transformation enhancement fusion algorithm, adaptive transformation processing is performed on the nonlinearly mapped feature data to obtain feature data after adaptive transformation, and fusion processing is performed on the feature data after adaptive transformation to form comprehensive multi-dimensional data.

[0012] Preferably, the S2 specifically includes: In the implementation process of the orthodontic treatment effect evaluation algorithm based on fuzzy reasoning, the comprehensive multi-dimensional data is input into the fuzzy reasoning network for fuzzy reasoning processing. In the fuzzy reasoning process, the membership value is generated according to the fuzzification degree of the comprehensive multi-dimensional data.

[0013] Preferably, the S2 specifically includes: In the implementation process of the orthodontic treatment effect evaluation algorithm based on fuzzy reasoning, based on the membership value of the comprehensive multi-dimensional data, the fuzzy reasoning rule is used for deduction to obtain a fuzzy reasoning result; and the fuzzy reasoning result is de-fuzzified to obtain a final treatment effect evaluation result.

[0014] Preferably, the S2 specifically includes: The final treatment effect evaluation result is compared with a set threshold value to obtain an evaluation result, and the orthodontic treatment effect is evaluated according to the evaluation result.

[0015] The beneficial effects of the technical solutions of the present application are: 1. The self-adaptive transformation enhancement fusion method based on similarity effectively improves the correlation between the preprocessed multi-dimensional data, enhances the fusion effect, and provides more accurate comprehensive multi-dimensional data, thereby reducing the influence of errors of a single data source on treatment effect evaluation.

[0016] 2. The orthodontic treatment effect evaluation algorithm based on fuzzy reasoning is introduced, the fuzzy reasoning network is used for treatment effect evaluation, the complex nonlinear relationship between the comprehensive multi-dimensional data can be processed, and a more accurate treatment effect evaluation result can be derived according to the membership values of the comprehensive multi-dimensional data. High-order defuzzification makes the final treatment effect evaluation result more accurate, can provide more scientific and reasonable treatment feedback, and supports the clinician to make better decisions during the treatment process. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of the orthodontic treatment effect evaluation method according to the present application. DETAILED DESCRIPTION

[0018] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined object of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0020] The specific scheme of the orthodontic treatment effect evaluation method provided by the present application will be specifically described below in conjunction with the drawings.

[0021] Referring to the drawings Figure 1 , which shows a flowchart of the orthodontic treatment effect evaluation method provided by an embodiment of the present application. The method comprises the following steps: S1. Real-time acquisition and preprocessing of multi-dimensional data to obtain preprocessed multi-dimensional data; the preprocessed multi-dimensional data is subjected to self-adaptive transformation enhancement fusion algorithm based on similarity for enhancement fusion processing to obtain comprehensive multi-dimensional data; Firstly, the image data of the patient is obtained through image acquisition equipment such as digital X-ray, CBCT, 3D oral scanner, etc.; the treatment record of the patient in the treatment process is obtained through the existing hospital information management system, including the treatment plan, treatment operation, appliances used in the treatment process (such as dental brace type, mechanical parameter, etc.) and treatment frequency of each visit, etc.; the feedback data of the patient is obtained through the feedback of equipment such as wearable device, including pain intensity, tooth activity, fitting condition of the treatment site, etc.; the multi-dimensional data of the patient in the orthodontic treatment process is obtained, and further, the multi-dimensional data is preprocessed to obtain preprocessed multi-dimensional data to ensure that different types of multi-dimensional data can be effectively fused; such as denoising and standardization processing of the image data using image denoising algorithms such as median filtering and mean filtering to obtain preprocessed image data; the treatment record is processed such as structured conversion and time sequencing to obtain preprocessed treatment record; the feedback data of the patient is quantitatively processed and time-synchronized to obtain preprocessed feedback data of the patient; and the preprocessed multi-dimensional data is normalized to ensure dimensional consistency. The above preprocessing process and normalization process are well-known technical means to those skilled in the art, and will not be described here.

[0022] Further, the preprocessed multi-dimensional data is enhanced and fused using an adaptive transformation enhancement fusion algorithm based on similarity, which describes the correlation of the preprocessed multi-dimensional data using the similarity between features, and then enhances and fuses the preprocessed multi-dimensional data through multi-scale adaptive transformation to obtain comprehensive multi-dimensional data, the specific implementation process is as follows: First, the preprocessed multi-dimensional data is preliminarily feature-extracted using existing feature engineering techniques to obtain initial feature data, such as preliminarily feature-extracting the preprocessed image data through a convolutional neural network to obtain initial feature data of the image such as tooth position, tooth movement, and fitting condition of the dental appliance; preliminarily feature-extracting the preprocessed treatment record and the preprocessed feedback data of the patient using a long short-term memory network to obtain initial feature data of the preprocessed treatment record and the preprocessed feedback data of the patient.

[0023] Further, based on the initial feature data, an adjacency matrix is constructed to represent the similarity between the initial feature data of each feature and further describe the correlation of the preprocessed multi-dimensional data by combining a distance-based similarity weighted adjacency matrix and a space-time decay term, and the calculation formula is as follows: , Wherein, is the initial feature data of the i-th feature and the initial feature data of the j-th feature. ​similarity of the i-th initial feature data is the i-th initial feature data . is the i-th initial feature data . is the j-th element of the i-th initial feature data . is the j-th element of the i-th initial feature data . is the j-th element of the i-th initial feature data . is the dimension of the feature is the i-th initial feature data . is the j-th element of the i-th initial feature data . is the j-th element of the i-th initial feature data is the total number of initial feature data is the L2 norm of the difference between the i-th initial feature data and the j-th initial feature data . is the L2 norm of the difference between the i-th initial feature data and the j-th initial feature data . is the L2 norm of the difference between the i-th initial feature data and the j-th initial feature data . and is a time decay term and are the timestamps of the i-th initial feature data and the j-th initial feature data, respectively . is a time decay factor for controlling the degree of influence of the time difference on the time similarity, and the specific value is determined according to the specific scene requirement . is the Euclidean distance between the i-th initial feature data and the j-th initial feature data . is the Euclidean distance between the i-th initial feature data and the j-th initial feature data . is the Euclidean distance between the i-th initial feature data and the j-th initial feature data . and is a space decay term is a space decay factor for controlling the influence of the spatial difference on the spatial similarity, and the specific value is determined according to the specific scene requirement based on the similarity between the initial feature data, each initial feature data is nonlinearly mapped using an existing nonlinear activation function to obtain nonlinearly mapped feature data .

[0024] Furthermore, existing adaptive transformation techniques, such as attention mechanisms, are used to adaptively transform the feature data after nonlinear mapping, achieving adaptive adjustment between features, improving the fusion effect, and obtaining feature data after adaptive transformation. .

[0025] After adaptive transformation, the transformed feature data are fused to form comprehensive multi-dimensional data. .

[0026] S2. The comprehensive multi-dimensional data is evaluated and analyzed using a fuzzy reasoning-based orthodontic treatment effect evaluation algorithm to obtain the evaluation results; based on the evaluation results, the orthodontic treatment effect is evaluated.

[0027] The orthodontic treatment effect evaluation algorithm based on fuzzy inference is used to evaluate and analyze the comprehensive multi-dimensional data to obtain the evaluation results. The orthodontic treatment effect evaluation algorithm based on fuzzy inference achieves the evaluation of orthodontic treatment effect by combining nonlinear inference mechanism and high-order defuzzification processing. The specific implementation process is as follows: First, comprehensive multi-dimensional data is input into a fuzzy inference network for fuzzy inference processing. This network is used to deduce the possible states of orthodontic treatment effects based on the fuzzy relationships between different comprehensive multi-dimensional data. To ensure high flexibility in the fuzzy inference process, a nonlinear inference mechanism based on fuzzy rules is designed, where each fuzzy rule infers based on the membership value of the comprehensive multi-dimensional data.

[0028] In fuzzy inference, membership values ​​are first generated based on the degree of fuzzification of the multi-dimensional data. Based on existing fuzzy control theory, fuzzy neural networks (FNNs) are derived and mixed with the logistic function and arctangent function to obtain the membership values, specifically in the following form: , in, It is the first Comprehensive multi-dimensional data In fuzzy sets The membership value in the fuzzy set is used to quantitatively represent the degree of membership of a feature in a certain fuzzy set. It is the first step in the fuzzy reasoning process. A comprehensive multi-dimensional data and the first The membership relationship between several fuzzy sets is a fuzzy set; It is the first A comprehensive, multi-dimensional dataset; It is the first Comprehensive multi-dimensional data The central value used in fuzzy inference is set based on expert experience, with a reference range of values. ; is the width or sensitivity of the membership function, determines the steepness of the membership function, is determined according to the expert experience method, and the reference value range is ; is a feature adjustment factor, used to control the influence strength of the first synthetic multi-dimensional data on the membership value, is determined according to the expert experience method, and the reference value range is ; is an arctangent function; indicates the comprehensive fuzzy membership corresponding to the nonlinear response strength of the first synthetic multi-dimensional data in approaching the ideal center state combined with its growth trend, reflecting the joint fuzzy compliance degree in numerical rationality and trend activity; The function of the above formula is to convert each synthetic multi-dimensional data into a membership value, which reflects the membership degree of the synthetic multi-dimensional data in the current fuzzy set. In this way, the fine-grained information of each synthetic multi-dimensional data can be preserved, and the nonlinear relationship between the features and the orthodontic treatment effect can also be captured.

[0029] According to the membership values of the synthetic multi-dimensional data, the possible states of the orthodontic treatment effect are derived through fuzzy reasoning rules. The design of the fuzzy reasoning rules considers the complex interaction between multi-dimensional features. Based on the existing fuzzy membership reasoning system, a feedforward disturbance modulation mechanism derived from non-standard control modeling theory is introduced, and the output result after fuzzy reasoning, i.e., the fuzzy reasoning result, is obtained. Next, the fuzzy reasoning result is de-fuzzied using existing de-fuzzification techniques such as centroid method, maximum membership method, and intermediate maximum method, to obtain the de-fuzzied result, i.e., the final treatment effect evaluation result .

[0030] Finally, the is compared with the threshold value determined according to the expert experience method, to obtain the specific evaluation result, i.e., when the is greater than or equal to the threshold value, it indicates that the treatment effect is good; when the is less than the threshold value, it indicates that the treatment effect is poor.

[0031] In summary, an evaluation method for orthodontic treatment effect is completed.

[0032] The order of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous.

[0033] The various embodiments in the specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the difference from other embodiments.

[0034] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for evaluating the effectiveness of orthodontic treatment, characterized in that, Includes the following steps: S1. Real-time acquisition and preprocessing of multi-dimensional data to obtain preprocessed multi-dimensional data; The preprocessed multidimensional data is enhanced and fused using a similarity-based adaptive transformation enhancement fusion algorithm to obtain comprehensive multidimensional data. S2. The comprehensive multi-dimensional data is evaluated and analyzed using a fuzzy reasoning-based orthodontic treatment effect evaluation algorithm to obtain the evaluation results; based on the evaluation results, the orthodontic treatment effect is evaluated.

2. The method for evaluating the effect of orthodontic treatment according to claim 1, characterized in that, S1 specifically includes: The similarity-based adaptive transformation enhancement fusion algorithm uses the similarity between features to describe the correlation between preprocessed multi-dimensional data, and then uses multi-scale adaptive transformation to enhance and fuse the preprocessed multi-dimensional data to obtain comprehensive multi-dimensional data.

3. The method for evaluating the effect of orthodontic treatment according to claim 2, characterized in that, S1 specifically includes: In the implementation of the similarity-based adaptive transformation enhancement fusion algorithm, feature extraction is performed on the preprocessed multi-dimensional data to obtain initial feature data.

4. The method for evaluating the effect of orthodontic treatment according to claim 3, characterized in that, S1 specifically includes: In the implementation of the similarity-based adaptive transformation enhancement fusion algorithm, based on the initial feature data, a similarity-weighted adjacency matrix based on distance and a spatiotemporal decay term are combined to construct an adjacency matrix representing the similarity between the initial feature data of each feature.

5. The method for evaluating the effect of orthodontic treatment according to claim 4, characterized in that, S1 specifically includes: In the implementation of the similarity-based adaptive transformation enhancement fusion algorithm, based on the similarity between the initial feature data, a nonlinear mapping is performed on each initial feature data to obtain the nonlinearly mapped feature data.

6. The method for evaluating the effect of orthodontic treatment according to claim 5, characterized in that, S1 specifically includes: In the implementation of the similarity-based adaptive transformation enhancement fusion algorithm, the feature data after nonlinear mapping is subjected to adaptive transformation processing to obtain the feature data after adaptive transformation; and the feature data after adaptive transformation is fused to form comprehensive multi-dimensional data.

7. The method for evaluating the effect of orthodontic treatment according to claim 1, characterized in that, S2 specifically includes: In the implementation of the orthodontic treatment effect evaluation algorithm based on fuzzy reasoning, comprehensive multi-dimensional data is input into the fuzzy reasoning network for fuzzy reasoning processing. During the fuzzy reasoning process, membership values ​​are generated according to the degree of fuzzification of the comprehensive multi-dimensional data.

8. The method for evaluating the effect of orthodontic treatment according to claim 7, characterized in that, S2 specifically includes: In the implementation of the orthodontic treatment effect evaluation algorithm based on fuzzy reasoning, the fuzzy reasoning result is obtained by using the membership value of comprehensive multi-dimensional data and fuzzy reasoning rules; and the fuzzy reasoning result is defuzzified to obtain the final treatment effect evaluation result.

9. The method for evaluating the effect of orthodontic treatment according to claim 8, characterized in that, S2 specifically includes: The final treatment effect assessment result is compared with the set threshold to obtain the assessment result, and the orthodontic treatment effect is evaluated based on the assessment result.

Images