Method and device for assessing caries risk, electronic device and storage medium

Abstract

The application relates to a caries risk assessment method and device, electronic equipment and a storage medium, wherein the method comprises the following steps: collecting patient information to be trained and caries risk data corresponding to the patient information to be trained; preprocessing the patient information to be trained and the caries risk data to obtain caries risk assessment standard data, and then establishing a training data set and a test data set for caries risk assessment; based on a preset logistic regression algorithm and a plurality of caries classification models, a caries risk detection initial model is constructed, the caries risk detection initial model is trained through the training data set, the trained caries risk detection initial model is fine-tuned by using the test data set, a caries risk detection model meeting a preset condition is generated, and in an online prediction stage, the caries risk detection model is used to perform a caries risk assessment operation. Therefore, the problems that the prior art cannot combine caries prevention and risk assessment, and it is difficult to assess and predict individual caries risk are solved.

Description

Technical Field

[0001] This application relates to the fields of data analysis and health management technology, and in particular to a method, device, electronic device and storage medium for assessing the risk of dental caries. Background Technology

[0002] With economic and social development, people's awareness of oral health and aesthetics is constantly improving, leading to an increasing demand for orthodontic treatment. Orthodontic treatment is a common method for clinically correcting malocclusion. However, orthodontic treatment alters the oral environment, increases the difficulty for patients to clean their teeth themselves, and significantly increases the risk of tooth decay.

[0003] Studies by scholars at home and abroad have found that 50%-80% of patients experience enamel demineralization after orthodontic treatment. If not controlled in time, this can further develop into tooth decay, seriously affecting the effectiveness of orthodontic treatment, reducing patient satisfaction with the orthodontic results, and severely damaging the aesthetics and health of teeth.

[0004] However, existing technologies cannot combine caries prevention with risk assessment, making it difficult to assess and predict individual caries risk, which urgently needs to be addressed. Summary of the Invention

[0005] This application provides a method, device, electronic device, and storage medium for assessing dental caries risk, in order to solve the problems that existing technologies cannot combine dental caries prevention with risk assessment, and that it is difficult to assess and predict individual dental caries risk.

[0006] The first aspect of this application provides a method for assessing dental caries risk, applied in an offline training phase, comprising the following steps: collecting patient information to be trained and corresponding dental caries risk data; preprocessing the patient information to be trained and the dental caries risk data to obtain dental caries risk assessment standard data, and establishing a training dataset and a test dataset for dental caries risk assessment based on the dental caries risk assessment standard data; constructing an initial dental caries risk detection model based on a preset logistic regression algorithm and multiple dental caries classification models, training the initial dental caries risk detection model using the training dataset, and fine-tuning the trained initial dental caries risk detection model using the test dataset to generate a dental caries risk detection model that meets preset conditions, so as to perform dental caries risk assessment operations using the dental caries risk detection model in the online prediction phase.

[0007] Optionally, in one embodiment of this application, the step of collecting the patient information to be trained and the corresponding caries risk data includes: determining multiple caries risk questions based on multiple preset caries risk factors, and constructing a caries risk questionnaire database through the multiple caries risk questions; determining the risk weight of each option corresponding to each caries risk question in the multiple caries risk questions, so as to obtain the risk score of each option through the risk weight of each option; and obtaining the caries risk data corresponding to the patient information to be trained based on the risk score of each option and the caries risk questionnaire database.

[0008] Optionally, in one embodiment of this application, the step of constructing an initial caries risk detection model based on a preset logistic regression algorithm and multiple caries classification models, and training the initial caries risk detection model using the training dataset, includes: constructing the multiple caries classification models, obtaining each caries risk feature corresponding to the training dataset, and inputting each caries risk feature into each of the multiple caries classification models to output the probability of all caries risk levels corresponding to each caries risk feature, wherein all caries risk levels include low-risk caries, medium-risk caries, and high-risk caries; and utilizing the logistic regression algorithm... A weighted regression operation is performed on the probabilities of all caries risk levels output by the multiple caries classification models to determine the final caries risk level corresponding to each caries risk feature; the weight of each caries risk feature is determined according to each caries classification model; a weighted average operation is performed on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature; the final weights of all caries risk features are sorted to generate a final weight ranking result, and the key risk features corresponding to the training dataset are determined according to the final weight ranking result, so as to train the initial caries risk detection model through the key risk features.

[0009] A second aspect of this application provides a method for assessing dental caries risk, applied in an online detection phase, comprising the following steps: collecting personal information of a target patient and obtaining dental caries risk information corresponding to the target patient based on a pre-set dental caries risk questionnaire database; performing data preprocessing operations on the personal information of the patient and the dental caries risk information to generate data to be predicted; inputting the data to be predicted into a pre-set dental caries risk detection model to output dental caries risk detection results, and generating a dental caries risk assessment report based on the dental caries risk detection results, wherein the dental caries risk detection model is obtained by training a pre-set initial dental caries risk detection model using a pre-set training dataset.

[0010] A third aspect of this application provides a dental caries risk assessment device for offline training, comprising: a first acquisition module for acquiring patient information to be trained and corresponding dental caries risk data; an establishment module for preprocessing the patient information to be trained and the dental caries risk data to obtain dental caries risk assessment standard data, and establishing a training dataset and a test dataset for dental caries risk assessment based on the dental caries risk assessment standard data; and a training module for constructing an initial dental caries risk detection model based on a preset logistic regression algorithm and multiple dental caries classification models, training the initial dental caries risk detection model using the training dataset, and fine-tuning the trained initial dental caries risk detection model using the test dataset to generate a dental caries risk detection model that meets preset conditions, so that in the online prediction stage, the dental caries risk detection model can be used to perform dental caries risk assessment operations.

[0011] Optionally, in one embodiment of this application, the first acquisition module includes: a first construction unit, configured to determine multiple caries risk issues based on multiple preset caries risk factors, and construct a caries risk questionnaire database through the multiple caries risk issues; a first determination unit, configured to determine the risk weight of each option corresponding to each caries risk issue in the multiple caries risk issues, so as to obtain the risk score of each option through the risk weight of each option; and an acquisition unit, configured to acquire caries risk data corresponding to the patient information to be trained based on the risk score of each option and the caries risk questionnaire database.

[0012] Optionally, in one embodiment of this application, the training module includes: a second construction unit, configured to construct the plurality of dental caries classification models, obtain each dental caries risk feature corresponding to the training dataset, and input each dental caries risk feature into each of the plurality of dental caries classification models to output the probability of all dental caries risk levels corresponding to each dental caries risk feature, wherein the all dental caries risk levels include low-risk dental caries, medium-risk dental caries, and high-risk dental caries; and a weighted regression unit, configured to use the logistic regression algorithm to weight the probabilities of all dental caries risk levels output by the plurality of dental caries classification models. The system employs a weighted regression operation to determine the final caries risk level corresponding to each caries risk feature; a second determination unit to determine the weight of each caries risk feature based on each caries classification model; a weighted averaging unit to perform a weighted average operation on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature; and a sorting unit to sort the final weights of all caries risk features, generate a final weight sorting result, and determine the key risk features corresponding to the training dataset based on the final weight sorting result, so as to train the initial caries risk detection model using the key risk features.

[0013] A fourth aspect of this application provides a dental caries risk assessment device applied in an online prediction stage, comprising: a second acquisition module for acquiring personal information of a target patient and obtaining dental caries risk information corresponding to the target patient based on a preset dental caries risk questionnaire database; a preprocessing unit for performing data preprocessing operations on the personal information of the patient and the dental caries risk information to generate data to be predicted; and an assessment unit for inputting the data to be predicted into a preset dental caries risk detection model to output dental caries risk detection results and generating a dental caries risk assessment report based on the dental caries risk detection results, wherein the dental caries risk detection model is obtained by training a preset initial dental caries risk detection model with a preset training dataset.

[0014] A fifth aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the caries risk assessment method as described in the above embodiments.

[0015] A sixth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for assessing the risk of dental caries.

[0016] Therefore, the embodiments of this application have the following beneficial effects:

[0017] The embodiments of this application can collect patient information to be trained and corresponding caries risk data; preprocess the patient information and caries risk data to obtain caries risk assessment standard data, and establish training and testing datasets for caries risk assessment based on the caries risk assessment standard data; construct an initial caries risk detection model based on a preset logistic regression algorithm and multiple caries classification models; train the initial caries risk detection model using the training dataset; and fine-tune the trained initial caries risk detection model using the testing dataset to generate a caries risk detection model that meets preset conditions, so as to perform caries risk assessment operations using the caries risk detection model in the online prediction stage. This application combines caries prevention with risk assessment, thereby carrying out multi-faceted and personalized caries management during orthodontic treatment, enhancing patients' awareness of caries prevention and control, reducing the incidence of caries, improving patients' oral health, and reducing the probability of enamel demineralization and caries during orthodontic treatment. Therefore, it solves the problems of existing technologies that cannot combine caries prevention with risk assessment and are difficult to assess and predict individual caries risk.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0020] Figure 1 A flowchart illustrating a method for assessing caries risk during offline training, according to an embodiment of this application;

[0021] Figure 2 A schematic diagram of a caries risk radar provided for one embodiment of this application;

[0022] Figure 3 A schematic diagram illustrating the execution logic of a method for assessing dental caries risk, provided as an embodiment of this application;

[0023] Figure 4 A flowchart illustrating a method for assessing dental caries risk in an online prediction phase, according to an embodiment of this application;

[0024] Figure 5 This is an example diagram of a caries risk assessment device applied during an offline training phase, according to an embodiment of this application.

[0025] Figure 6 This is an example diagram of a caries risk assessment device applied in the online prediction stage according to an embodiment of this application;

[0026] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0027] Among them, 10-a caries risk assessment device applied to the offline training stage, 20-a caries risk assessment device applied to the online prediction stage; 101-first acquisition module, 102-establishment module, 103-training module; 201-second acquisition module, 202-preprocessing module, 203-evaluation module; 701-memory, 702-processor, 703-communication interface. Detailed Implementation

[0028] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0029] The following description, with reference to the accompanying drawings, describes a method, apparatus, electronic device, and storage medium for assessing dental caries risk according to embodiments of this application. Addressing the problems mentioned in the background section, this application provides a method for assessing dental caries risk. In this method, information about a patient to be trained and corresponding dental caries risk data are collected. The patient information and dental caries risk data are preprocessed to obtain standard data for dental caries risk assessment. A training dataset and a test dataset for dental caries risk assessment are established based on the standard data. An initial dental caries risk detection model is constructed based on a preset logistic regression algorithm and multiple dental caries classification models. This initial model is trained using the training dataset and fine-tuned using the test dataset to generate a dental caries risk detection model that meets preset conditions. During the online prediction phase, the dental caries risk detection model is used to perform dental caries risk assessment. This application combines dental caries prevention with risk assessment, thereby enabling multi-faceted and personalized dental caries management during orthodontic treatment. This enhances patients' awareness of dental caries prevention and control, reduces the incidence of dental caries, improves patients' oral health, and reduces the probability of enamel demineralization and dental caries during orthodontic treatment. This solves the problems that existing technologies cannot combine caries prevention with risk assessment, and it is difficult to assess and predict individual caries risk.

[0030] Specifically, Figure 1 A flowchart illustrating a method for assessing caries risk during the offline training phase, provided as an embodiment of this application.

[0031] like Figure 1 As shown, the method for assessing the risk of dental caries includes the following steps:

[0032] In step S101, information on the patient to be trained and the corresponding caries risk data are collected.

[0033] The embodiments of this application can first collect information on patients to be trained and corresponding caries risk-related data through questionnaires, oral examination reports, medical records, etc., thereby providing data basis for training the initial model for subsequent caries risk detection.

[0034] Optionally, in one embodiment of this application, collecting patient information to be trained and corresponding caries risk data includes: determining multiple caries risk questions based on multiple preset caries risk factors, and constructing a caries risk questionnaire database through the multiple caries risk questions; determining the risk weight of each option corresponding to each caries risk question in the multiple caries risk questions, so as to obtain the risk score of each option through the risk weight of each option; and obtaining caries risk data corresponding to the patient information to be trained based on the risk score of each option, the caries risk questionnaire database, and the patient information to be trained.

[0035] It should be noted that the embodiments of this application can pre-set a caries risk questionnaire database containing multiple caries risk factors. Each caries risk factor contains multiple caries risk questions. The caries risk questionnaire database includes three types of questionnaires: a caries risk grading assessment questionnaire for people who want to undergo invisible orthodontic treatment, a caries risk assessment questionnaire for people who are currently wearing invisible orthodontic appliances, and a caries risk assessment questionnaire for people who have completed their invisible orthodontic treatment.

[0036] Furthermore, embodiments of this application assign different risk weights to different options in each caries risk question, and calculate the risk score of each option based on the risk weights; doctors can select caries risk questionnaires according to the patient's condition to collect patient information and caries risk-related data.

[0037] Specifically, risk factors for dental caries include: genetic and environmental factors, dietary and lifestyle factors, saliva analysis factors, dental examination factors, and general physical condition factors, such as... Figure 2 As shown.

[0038] In practice, genetic and environmental factors accounted for 4% of the caries risk questionnaire; dietary and lifestyle factors accounted for 36%; saliva quality testing factors accounted for 12%; dental examination factors accounted for 36%; and general physical condition factors accounted for 12%.

[0039] Among these, genetic and environmental factors include: the severity of dental caries (cavities) in immediate family members;

[0040] Dietary and lifestyle factors include: frequency of consuming sugary foods and drinks, consumption of sugary foods and drinks while wearing clear aligners, whether teeth are cleaned after eating and drinking sugary drinks, whether clear aligners are cleaned after eating and drinking sugary drinks, the method of cleaning teeth after eating and drinking sugary drinks, whether dental floss or other oral hygiene aids are used, the method of cleaning clear aligners, whether fluoride is used, and whether plaque disclosing agents have been used.

[0041] Factors affecting saliva quality include: saliva pH, saliva buffering capacity, and saliva flow rate.

[0042] Factors to consider in a dental examination include: the presence of chalky spots on the tooth surface before orthodontic treatment; the presence of new chalky spots on the tooth surface during orthodontic treatment; the presence of caries or fillings (including those with cavities and those without) confirmed by visual inspection or radiographic examination; whether there has been tooth loss due to caries in the past 36 months; the presence of visible plaque on the tooth surface; the presence of abnormal tooth morphology (which can lead to inadequate tooth cleaning); ≥1 caries and filling on the proximal surfaces of the teeth; whether there is root exposure; and the presence of defective restorations in the mouth (restorations with overhangs and / or open margins, food impaction on the proximal surfaces of the restorations).

[0043] General physical factors include: the reason for undergoing invisible orthodontic treatment and the number of months the invisible orthodontic appliances have been worn.

[0044] In the embodiments of this application, a total of 25 questions are set for the risk of dental caries, including 1 question on genetic and environmental factors, 9 questions on dietary and lifestyle factors, 3 questions on general physical condition factors, 3 questions on saliva condition testing factors, and 9 questions on dental specialist examination factors.

[0045] In practice, the caries risk detection report includes each caries risk factor and its corresponding risk score, the caries risk visualization results, the caries risk level and coping measures, and the caries risk level includes low risk, medium risk or high risk (i.e. low risk caries, medium risk caries or high risk caries).

[0046] Low risk includes: good oral health, no obvious signs of tooth decay, good oral hygiene habits, healthy eating habits, and normal oral pH level;

[0047] Prevention measures: Regular dental checkups and cleanings, and maintaining good oral hygiene habits, including brushing, flossing, and rinsing. Pay attention to a healthy diet, reduce sugar intake, eat more fruits and vegetables, and avoid excessively acidic foods and drinks. Have regular dental checkups and treat any signs of cavities promptly.

[0048] Medium risk: Oral health is generally good, with minor signs of tooth decay, oral hygiene habits are average, dietary habits have room for improvement, and oral pH is low.

[0049] Countermeasures: Strengthen oral hygiene habits, including regular deep cleaning, using mouthwash, and regular dental checkups; adjust your diet, reducing sugar and carbonated beverage intake and increasing fruit and vegetable intake. Regularly check the progression of cavities and seek timely treatment.

[0050] High risk: Poor oral health, obvious signs of tooth decay, poor oral hygiene habits, unhealthy diet, and significantly low oral pH.

[0051] Prevention measures: 1. Brush your teeth 2-3 times a day with fluoride toothpaste; 2. Rinse your mouth with fluoride mouthwash after meals; 3. Use dental floss and other auxiliary tools; 4. Get fluoride varnish or fluoride foam treatment every 3 months; 5. Monitor for dental caries every 3 months; 6. For patients wearing invisible aligners, it is recommended to use plaque revealing agents and aligner cleaning products to help clean the invisible aligners; 7. It is recommended to use remineralizing gel.

[0052] Each option in the question is assigned a different risk weight based on its impact on the risk of dental caries. Patients can select the appropriate option according to their own situation, and then calculate the risk score for each question based on the weight. Finally, the overall dental caries risk assessment result is obtained by combining the results, as shown in Table 1.

[0053] Table 1

[0054]

[0055] In a specific embodiment of the present invention, the calculation process and formulas for each score are shown in the rightmost column of Table 1. The caries risk questionnaire contains multiple caries risk factors, each caries risk factor contains multiple questions, and each question contains multiple options, with a maximum of 3 options. Based on the patient's questionnaire results, the number of options (i.e., the number of A, B, and C) corresponding to each caries risk factor is obtained. Table 1 is then consulted to calculate the score corresponding to each caries risk factor.

[0056] For example, questions 1, 2, and 25 reflect "general physical condition." The answers to question 1 include "A, B," the answers to question 2 include "B, C," and the answers to question 25 include "A, B." Therefore, the answers to questions reflecting "general physical condition" include 2 "A," 3 "B," and 1 "C," as shown in Table 1. A patient's answers, as shown in Table 2, include one "B," one "C," and one "A." According to the calculation method in Table 1, the patient's answer contains one "C," which conforms to the calculation method "A = 0 - 2, B = 1 - 3, C = 1, 80 - 100," resulting in a score of 80 + 1 / 1 * 19 = 99.

[0057] The questionnaire survey results of a patient collected in this embodiment are shown in Table 2. Based on the data in Table 2 and Table 1, the comprehensive scoring results are shown in Table 3. The score for genetic and environmental factors is 41, the score for dietary and lifestyle factors is 73, the score for saliva quality testing factors is 79, the score for oral specialist examination factors is 86, and the score for general physical condition factors is 99. A caries risk radar chart is obtained based on the scores, as shown below. Figure 2 As shown, the final caries risk level is high, and the following coping measures are provided to the patient:

[0058] 1. Brush your teeth 2-3 times a day using fluoride toothpaste;

[0059] 2. Rinse your mouth with fluoride mouthwash after meals;

[0060] 3. Dental floss and other auxiliary tools are required;

[0061] 4. Apply fluorinated protective varnish or fluorinated foam treatment every 3 months;

[0062] 5. Monitor oral caries every 3 months;

[0063] 6. For patients wearing clear aligners, it is recommended to use plaque disclosing agents and aligner cleaning products to help clean the clear aligners.

[0064] 7. It is recommended to use remineralized gel.

[0065] Table 2

[0066]

[0067] Table 3

[0068]

[0069]

[0070] Therefore, the embodiments of this application effectively ensure the execution of subsequent operations such as the construction and training of the initial model for caries risk detection by collecting patient information and caries risk-related data.

[0071] In step S102, the patient information and caries risk data to be trained are preprocessed to obtain caries risk assessment standard data, and a training dataset and a test dataset for caries risk assessment are established based on the caries risk assessment standard data.

[0072] In step S103, based on the preset logistic regression algorithm and multiple caries classification models, an initial caries risk detection model is constructed. The initial caries risk detection model is trained using a training dataset, and the trained initial caries risk detection model is fine-tuned using a test dataset to generate a caries risk detection model that meets preset conditions. In the online prediction stage, the caries risk detection model is used to perform caries risk assessment.

[0073] Furthermore, embodiments of this application also require preprocessing operations such as data deduplication, missing value handling, and outlier handling of the patient information and caries risk data to obtain caries risk assessment standard data. The caries risk assessment standard data is then divided into a training set and a test set according to a certain ratio to ensure the accuracy and completeness of the data. Subsequently, embodiments of this application can use a logistic regression algorithm to construct an initial caries risk detection model. This initial model is then trained using the training set to obtain the trained initial caries risk detection model. Finally, a test set is selected to test and evaluate the trained initial caries risk detection model, thereby obtaining the final caries risk detection model.

[0074] Optionally, in one embodiment of this application, an initial caries risk detection model is constructed based on a preset logistic regression algorithm and multiple caries classification models. The initial caries risk detection model is trained using a training dataset, including: constructing multiple caries classification models, obtaining each caries risk feature corresponding to the training dataset, and inputting each caries risk feature into each of the multiple caries classification models to output the probability of all caries risk levels corresponding to each caries risk feature, wherein all caries risk levels include low-risk caries, medium-risk caries, and high-risk caries; using a logistic regression algorithm to perform a weighted regression operation on the probabilities of all caries risk levels output by the multiple caries classification models to determine the final caries risk level corresponding to each caries risk feature; determining the weight of each caries risk feature according to each caries classification model; performing a weighted average operation on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature; sorting the final weights of all caries risk features to generate a final weight ranking result, and determining the key risk features corresponding to the training dataset based on the final weight ranking result, so as to train the initial caries risk detection model using the key risk features.

[0075] Specifically, in the embodiments of this application, a caries classification model can be constructed using a random forest model, a gradient boosting decision tree, and an adaptive boosting algorithm, respectively. The caries risk features are used as input to predict the probability of all caries risk levels caused by each caries risk feature using each caries classification model.

[0076] Secondly, the embodiments of this application can determine the probability of all caries risk levels caused by each caries risk feature based on caries risk-related data and information, using each caries classification model; and use a logistic regression algorithm to perform weighted regression on the probabilities of all caries risk levels output by each caries classification model, so as to determine the caries risk level through the output results of the weighted regression. This caries risk level takes into account the characteristics of multiple classification models, avoids the problem of the limitations of a single classification model, improves the accuracy of determining the caries risk level, and thus can effectively determine the patient's risk of caries.

[0077] Furthermore, embodiments of this application also require determining the weight of each caries risk feature based on each caries classification model, and performing a weighted average of multiple weights for each caries risk feature to determine the final weight of each caries risk feature; subsequently, embodiments of this application require sorting the final weights of all caries risk features to generate a final weight ranking result, and taking the caries risk feature with the highest final weight as the key risk feature.

[0078] It should be noted that in the embodiments of this application, multiple caries classification models can be regarded as highly complex nonlinear feature converters. Patient caries data are fed into each caries classification model, and each caries classification model outputs a vector. Each value in the vector represents the probability of each caries risk level. The three vectors output by multiple caries classification models represent the perspectives of three different methods. By concatenating the three vectors, a high-dimensional representation of the original features can be obtained.

[0079] In practice, each caries classification model is an ensemble model based on decision trees. The decision tree can determine the importance of caries risk features by calculating the normalized value of the reduction in information entropy. The importance of each caries risk feature is extracted from each caries classification model as a weight. The average weight of the caries risk features in each caries classification model is calculated, and they are arranged from low to high according to the average weight to determine their influence level. The higher the value, the greater the influence. The caries risk features of the highest level are the key risk features.

[0080] After obtaining the key risk characteristics, the embodiments of this application further require sensitivity analysis to quantify the severity of the caries risk caused by the key risk characteristics.

[0081] Specifically, in an embodiment of this application, one key risk feature may be perturbed at a time while other caries risk features remain unchanged. The degree of change in classification accuracy on the verification dataset is then observed. The perturbation range for the caries risk feature ranges from an increase of 1 unit to an increase of 10 units, where each unit refers to one-tenth of the average value of all samples for that caries risk feature. During the perturbation of the caries risk feature, the greater the impact on the caries risk level classification, the more sensitive the caries risk feature is, i.e., the greater its influence on the severity of the caries risk.

[0082] In the specific implementation process, the embodiments of this application can use the Sigmoid function to map the output result after weighted regression to between 0 and 1, and determine the caries risk level with the highest probability as the final caries risk level result.

[0083] Finally, embodiments of this application can deploy the caries risk detection model into a caries management platform to facilitate the assessment and prediction of individual caries risk.

[0084] It should be noted that the caries management platform in this application embodiment mainly includes: a doctor subsystem, a user subsystem, a backend subsystem, and a blockchain platform; in the caries management platform, the backend subsystem is connected to the doctor subsystem and the user subsystem respectively, the doctor subsystem is connected to the user subsystem respectively, and the blockchain platform is connected to the doctor subsystem, the user subsystem, and the backend subsystem respectively.

[0085] The backend subsystem mainly includes the following modules:

[0086] The user management module is used to store user information on the blockchain platform, and the blockchain platform synchronizes all user information to all nodes on the blockchain platform.

[0087] The doctor management module is used to store doctor information on the blockchain platform, and the blockchain platform synchronizes all doctor information to all nodes on the blockchain platform.

[0088] The administrator module allows backend administrators to access the blockchain platform and retrieve all operation records of the administrator and the doctors they manage, as well as all user information of the managed doctors.

[0089] The questionnaire management module is used to store caries risk questionnaires on the blockchain platform; and the blockchain platform synchronizes caries risk questionnaires to all nodes on the blockchain platform.

[0090] The test report management module is used to store the generated caries risk test reports to the blockchain platform, and the blockchain platform synchronizes the caries risk test reports to all nodes on the blockchain platform.

[0091] The doctor subsystem mainly includes the following modules:

[0092] The doctor QR code module is used to generate a unique QR code for each doctor, which contains the doctor's personal information.

[0093] The doctor diagnosis storage module is used by doctors to store the medical record information of their associated users on the blockchain platform, and the blockchain platform synchronizes the medical record information to all nodes on the blockchain platform.

[0094] The doctor association module is used to receive association requests sent by users to the doctor, and after the doctor confirms the association, it stores the confirmed association information in the blockchain platform, and the blockchain platform synchronizes the association information to all nodes of the blockchain platform.

[0095] The doctor search module is used to retrieve all associated information about a doctor and the corresponding user information for each associated information from the blockchain platform.

[0096] The user subsystem mainly includes the following modules:

[0097] The user retrieval module is used to retrieve the doctor information that can be associated with the user's operation permissions and all associated information of the user in the blockchain platform.

[0098] The user association module is used to send an association request to the doctor subsystem based on the doctor selected by the user, and after receiving the doctor's confirmation of association information, store the confirmed association information in the blockchain platform, and synchronize the association information to all nodes of the blockchain platform.

[0099] After registering and logging in, users can directly access the correction module, which mainly includes the preparation unit, the correction in progress unit, the correction completion unit, and the doctor association unit.

[0100] The preparation unit obtains the caries risk assessment questionnaire for people who want to undergo invisible orthodontic treatment from the questionnaire management module and provides it to the patients. After the patients complete the questionnaire in turn, the results are transmitted to the caries risk detection model to obtain the caries risk level corresponding to each caries risk factor. The test results are viewed from the test list and displayed visually. At the same time, the test results are sent to the doctor, who makes a diagnosis based on the test results.

[0101] The orthodontic unit retrieves a caries risk assessment questionnaire for people wearing invisible aligners from the questionnaire management module and provides it to the patients. After the patients complete the questionnaire, the results are transmitted to the caries risk detection model to obtain the caries risk level corresponding to each caries risk factor. The test results can be viewed from the test list and visualized. At the same time, the test results are sent to the doctor, who makes a diagnosis based on the test results.

[0102] The orthodontic completion unit retrieves a caries risk assessment questionnaire for patients who have completed treatment with invisible aligners from the questionnaire management module and provides it to the patients. After the patients complete the questionnaire in turn, the results are transmitted to the caries risk detection model to obtain the caries risk level corresponding to each caries risk factor. The test results can be viewed from the test list and visualized. At the same time, the test results are sent to the doctor, who makes a diagnosis based on the test results.

[0103] The "Associate Doctor" section allows users to view doctors' personal information, including their name and contact information. Users can also choose to change doctors if they are dissatisfied with the service. To change doctors, users can search for the desired doctor by entering their name or scan the doctor's QR code to replace the original doctor's information. Each doctor can be associated with at least one user, and each user can be associated with one or more doctors.

[0104] The blockchain platform mainly includes the following modules:

[0105] The key management module is used to store the account information of the backend administrator, the doctor's account, and the user's account information, verify the identity of the user accessing the blockchain platform, and grant corresponding operation permissions to the verified user.

[0106] The node status management module is used to manage the operation status information of backend administrators, doctors, and users stored in the blockchain platform;

[0107] The smart contract module is used to store the associated information sent by the doctor subsystem;

[0108] The timestamp module is used to assign timestamps to all information stored in the blockchain platform;

[0109] The blockchain information monitoring module is used to monitor all information on the blockchain platform, the operation status information of the backend administrator, the operation status information of the doctor, and the operation status information of the user, and to provide real-time viewing of transaction information that occurs on the blockchain platform.

[0110] The execution logic of the caries risk assessment method of this application will be described below through a specific embodiment and in conjunction with the accompanying drawings.

[0111] Figure 3 This is a schematic diagram illustrating the execution logic of the caries risk assessment method described in this application. Figure 3 As shown, the procedure for assessing the risk of dental caries in this application is as follows:

[0112] S301: Collect patient information and data related to caries risk;

[0113] S302: Clean, filter, and encode the collected data, and divide it into training and test sets;

[0114] S303: Construct an initial model for caries risk detection based on logistic regression algorithm, and train the initial model for caries risk detection using the training set to obtain the trained initial model for caries risk detection;

[0115] S304: Select a test set to test and evaluate the trained initial caries risk detection model in order to obtain the final caries risk detection model;

[0116] S305: Input patient information and caries risk-related data into the trained caries risk detection model to generate a caries risk detection report and display it visually.

[0117] In summary, this application collects patient information and caries risk-related data, cleans, filters, and encodes the collected data; constructs a caries risk detection model based on a logistic regression algorithm, trains the model using a training set, and then tests and evaluates the trained model using a test set. Therefore, this application combines caries prevention with risk assessment and provides comprehensive, personalized caries management during orthodontic treatment, enhancing patients' awareness of caries prevention and control, reducing the incidence of caries, and improving patients' oral health.

[0118] The caries risk assessment method proposed in this application for offline training phase involves collecting patient information and corresponding caries risk data. The patient information and caries risk data are preprocessed to obtain caries risk assessment standard data. Training and testing datasets for caries risk assessment are then established based on this standard data. An initial caries risk detection model is constructed based on a preset logistic regression algorithm and multiple caries classification models. This initial model is trained using the training dataset and fine-tuned using the testing dataset to generate a caries risk detection model that meets preset conditions. This model is then used to perform caries risk assessment during the online prediction phase. This application combines caries prevention with risk assessment, enabling comprehensive and personalized caries management during orthodontic treatment. This enhances patients' awareness of caries prevention, reduces the incidence of caries, improves patients' oral health, and decreases the probability of enamel demineralization and caries during orthodontic treatment.

[0119] Figure 4 A flowchart illustrating a method for assessing caries risk in the online prediction phase, provided as an embodiment of this application.

[0120] like Figure 4 As shown, the method for assessing the risk of dental caries includes the following steps:

[0121] In step S401, the patient's personal information is collected, and the corresponding caries risk information of the target patient is obtained based on the preset caries risk questionnaire database.

[0122] In step S402, data preprocessing operations are performed on the patient's personal information and caries risk information to generate data to be predicted.

[0123] In step S403, the data to be predicted is input into the preset caries risk detection model to output the caries risk detection result, and a caries risk assessment report is generated based on the caries risk detection result. The caries risk detection model is obtained by training the preset caries risk detection initial model with the preset training dataset.

[0124] The caries risk assessment method proposed in this application for online prediction involves collecting personal information of target patients and obtaining corresponding caries risk information based on a pre-set caries risk questionnaire database. Data preprocessing is performed on the patient's personal information and caries risk information to generate data to be predicted. This data is then input into a pre-set caries risk detection model to output caries risk detection results, and a caries risk assessment report is generated based on these results. The caries risk detection model is obtained by training a pre-set initial caries risk detection model using a pre-set training dataset. This application combines caries prevention with risk assessment to conduct multi-faceted, personalized caries management during orthodontic treatment, enhancing patients' awareness of caries prevention and control, reducing the incidence of caries, improving patients' oral health, and decreasing the probability of enamel demineralization and caries during orthodontic treatment.

[0125] Secondly, the caries risk assessment device proposed according to the embodiments of this application is described with reference to the accompanying drawings.

[0126] Figure 5 This is a block diagram of a caries risk assessment device applied to the offline training phase according to an embodiment of this application.

[0127] like Figure 5 As shown, the caries risk assessment device 10 includes: a first acquisition module 101, an establishment module 102, and a training module 103.

[0128] The first acquisition module 101 is used to acquire information about the patient to be trained and the corresponding dental caries risk data.

[0129] Module 102 is established to preprocess patient information and caries risk data to obtain caries risk assessment standard data, and to establish training and test datasets for caries risk assessment based on the caries risk assessment standard data.

[0130] Training module 103 is used to construct an initial caries risk detection model based on a preset logistic regression algorithm and multiple caries classification models. The initial caries risk detection model is trained using a training dataset and fine-tuned using a test dataset to generate a caries risk detection model that meets preset conditions. In the online prediction stage, the caries risk detection model is used to perform caries risk assessment.

[0131] Optionally, in one embodiment of this application, the first acquisition module 101 includes: a first construction unit, a first determination unit, and an acquisition unit.

[0132] The first construction unit is used to identify multiple caries risk issues based on multiple preset caries risk factors, and to build a caries risk questionnaire database through multiple caries risk issues.

[0133] The first determining unit is used to determine the risk weight of each option corresponding to each caries risk question in multiple caries risk questions, so as to obtain the risk score of each option through the risk weight of each option.

[0134] The acquisition unit is used to obtain the caries risk data corresponding to the information of the patients to be trained based on the risk score of each option and the caries risk questionnaire database.

[0135] Optionally, in one embodiment of this application, the training module 103 includes: a second construction unit, a weighted regression unit, a second determination unit, a weighted average unit, and a sorting unit.

[0136] The second building unit is used to build multiple caries classification models, obtain each caries risk feature corresponding to the training dataset, and input each caries risk feature into each of the multiple caries classification models to output the probability of all caries risk levels corresponding to each caries risk feature. All caries risk levels include low-risk caries, medium-risk caries, and high-risk caries.

[0137] The weighted regression unit is used to perform a weighted regression operation on the probabilities of all caries risk levels output by multiple caries classification models using the logistic regression algorithm, in order to determine the final caries risk level corresponding to each caries risk feature.

[0138] The second determining unit is used to determine the weight of each caries risk feature based on each caries classification model.

[0139] The weighted average unit is used to perform a weighted average operation on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature.

[0140] The sorting unit is used to sort the final weights of all caries risk features, generate the final weight sorting result, and determine the key risk features corresponding to the training dataset based on the final weight sorting result, so as to train the initial caries risk detection model through the key risk features.

[0141] The dental caries risk assessment device proposed in this application for offline training includes a first acquisition module for acquiring patient information and corresponding dental caries risk data; a setup module for preprocessing the patient information and dental caries risk data to obtain dental caries risk assessment standard data, and establishing a training dataset and a test dataset for dental caries risk assessment based on the dental caries risk assessment standard data; and a training module for constructing an initial dental caries risk detection model based on a preset logistic regression algorithm and multiple dental caries classification models, training the initial dental caries risk detection model using the training dataset, and fine-tuning the trained initial dental caries risk detection model using the test dataset to generate a dental caries risk detection model that meets preset conditions, so that dental caries risk assessment can be performed using the dental caries risk detection model in the online prediction stage. This application combines dental caries prevention with risk assessment, thereby carrying out multi-faceted and personalized dental caries management during orthodontic treatment, enhancing patients' awareness of dental caries prevention and control, reducing the incidence of dental caries, improving patients' oral health, and reducing the probability of enamel demineralization and dental caries during orthodontic treatment.

[0142] Figure 6 This is a block diagram of a caries risk assessment device applied in the online pre-stage according to an embodiment of this application.

[0143] like Figure 6 As shown, the caries risk assessment device 20 applied in the online prediction stage includes: a second acquisition module 201, a preprocessing module 202, and an assessment module 203.

[0144] The second collection module 201 is used to collect the personal information of the target patient and obtain the corresponding caries risk information of the target patient based on a preset caries risk questionnaire database.

[0145] The preprocessing unit 202 is used to perform data preprocessing operations on patient personal information and caries risk information to generate data to be predicted.

[0146] Evaluation unit 203 is used to input the data to be predicted into a preset caries risk detection model to output caries risk detection results and generate a caries risk assessment report based on the caries risk detection results. The caries risk detection model is obtained by training a preset caries risk detection initial model with a preset training dataset.

[0147] It should be noted that the explanation of the aforementioned method embodiment for assessing caries risk also applies to the caries risk assessment device of this embodiment, and will not be repeated here.

[0148] The caries risk assessment device proposed in this application for online prediction includes a second acquisition module for collecting personal information of the target patient and obtaining caries risk information corresponding to the target patient based on a preset caries risk questionnaire database; a preprocessing unit for performing data preprocessing operations on the patient's personal information and caries risk information to generate data to be predicted; and an assessment unit for inputting the data to be predicted into a preset caries risk detection model to output caries risk detection results and generate a caries risk assessment report based on the caries risk detection results. The caries risk detection model is obtained by training a preset initial caries risk detection model using a preset training dataset. This application combines caries prevention with risk assessment to conduct multi-faceted and personalized caries management during orthodontic treatment, enhancing patients' awareness of caries prevention and control, reducing the incidence of caries, improving patients' oral health, and reducing the probability of enamel demineralization and caries during orthodontic treatment.

[0149] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include:

[0150] The memory 701, the processor 702, and the computer program stored on the memory 701 and capable of running on the processor 702.

[0151] When the processor 702 executes the program, it implements the caries risk assessment method provided in the above embodiments.

[0152] Furthermore, electronic devices also include:

[0153] Communication interface 703 is used for communication between memory 701 and processor 702.

[0154] The memory 701 is used to store computer programs that can run on the processor 702.

[0155] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0156] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0157] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0158] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0159] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for assessing the risk of dental caries.

[0160] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0161] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0162] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0163] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0164] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0165] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0166] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0167] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for assessing the risk of dental caries, characterized in that, When applied to the offline training phase, the following steps are included: Collect information on patients to be trained and corresponding caries risk data for those patients. The patient information to be trained and the caries risk data are preprocessed to obtain caries risk assessment standard data, and a training dataset and a test dataset for caries risk assessment are established based on the caries risk assessment standard data. Based on a preset logistic regression algorithm and multiple caries classification models, an initial caries risk detection model is constructed. The initial caries risk detection model is trained using the training dataset, and the trained initial caries risk detection model is fine-tuned using the test dataset to generate a caries risk detection model that meets preset conditions. In the online prediction stage, the caries risk detection model is used to perform caries risk assessment. The step of constructing an initial caries risk detection model based on a preset logistic regression algorithm and multiple caries classification models, and training the initial caries risk detection model using the training dataset, includes: Construct the multiple caries classification models, obtain each caries risk feature corresponding to the training dataset, and input each caries risk feature into each of the multiple caries classification models to output the probability of all caries risk levels corresponding to each caries risk feature, wherein all caries risk levels include low-risk caries, medium-risk caries, and high-risk caries; The logistic regression algorithm is used to perform a weighted regression operation on the probabilities of all caries risk levels output by the multiple caries classification models to determine the final caries risk level corresponding to each caries risk feature; The weight of each caries risk feature is determined based on each caries classification model. A weighted average operation is performed on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature; The final weights of all caries risk features are sorted to generate a final weight sorting result. The key risk features corresponding to the training dataset are determined based on the final weight sorting result, so as to train the initial caries risk detection model through the key risk features. Among them, a caries classification model is constructed by random forest model, gradient boosting decision tree and adaptive boosting algorithm respectively. The caries risk feature is used as input to predict the probability of all caries risk levels caused by each caries risk feature using each caries classification model. Sensitivity analysis was performed on the key risk features obtained. Each time, one key risk feature was perturbed while other caries risk features remained unchanged. The degree of change in classification accuracy on the verification dataset was observed.

2. The method according to claim 1, characterized in that, The collection of patient information for training and corresponding dental caries risk data includes: Based on multiple preset caries risk factors, multiple caries risk questions are identified, and a caries risk questionnaire database is constructed using these multiple caries risk questions; Determine the risk weight of each option corresponding to each of the multiple caries risk questions, so as to obtain the risk score of each option through the risk weight of each option; Based on the risk score of each option and the caries risk questionnaire database, obtain the caries risk data corresponding to the patient information to be trained.

3. A method for assessing the risk of dental caries, characterized in that, The caries risk assessment method described in claim 1 or 2, applied to the offline training phase, is applied to the online prediction phase, comprising the following steps: Collect the personal information of the target patients and obtain the corresponding caries risk information of the target patients based on the pre-set caries risk questionnaire database; Data preprocessing operations are performed on the patient's personal information and the caries risk information to generate data to be predicted; The data to be predicted is input into a preset caries risk detection model to output caries risk detection results, and a caries risk assessment report is generated based on the caries risk detection results. The caries risk detection model is obtained by training a preset caries risk detection initial model with a preset training dataset.

4. A caries risk assessment device, used in the offline training phase, characterized in that, The apparatus employs the caries risk assessment method as described in claim 1 or 2 for use in offline training phases, comprising: The first acquisition module is used to acquire information about the patient to be trained and the caries risk data corresponding to the patient information; A module is established to preprocess the patient information to be trained and the caries risk data to obtain caries risk assessment standard data, and to establish a training dataset and a test dataset for caries risk assessment based on the caries risk assessment standard data. The training module is used to construct an initial caries risk detection model based on a preset logistic regression algorithm and multiple caries classification models. The initial caries risk detection model is trained using the training dataset, and the trained initial caries risk detection model is fine-tuned using the test dataset to generate a caries risk detection model that meets preset conditions. In the online prediction stage, the caries risk detection model is used to perform caries risk assessment. The training module includes: The second construction unit is used to construct the plurality of caries classification models, obtain each caries risk feature corresponding to the training dataset, and input each caries risk feature into each of the plurality of caries classification models to output the probability of all caries risk levels corresponding to each caries risk feature, wherein all caries risk levels include low-risk caries, medium-risk caries and high-risk caries; The weighted regression unit is used to perform a weighted regression operation on the probabilities of all caries risk levels output by the multiple caries classification models using the logistic regression algorithm, so as to determine the final caries risk level corresponding to each caries risk feature; The second determining unit is used to determine the weight of each caries risk feature based on each caries classification model; The weighted average unit is used to perform a weighted average operation on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature; The sorting unit is used to sort the final weights of all caries risk features, generate the final weight sorting result, and determine the key risk features corresponding to the training dataset based on the final weight sorting result, so as to train the initial caries risk detection model through the key risk features. Among them, a caries classification model is constructed by random forest model, gradient boosting decision tree and adaptive boosting algorithm respectively. The caries risk feature is used as input to predict the probability of all caries risk levels caused by each caries risk feature using each caries classification model. Sensitivity analysis was performed on the key risk features obtained. Each time, one key risk feature was perturbed while other caries risk features remained unchanged. The degree of change in classification accuracy on the verification dataset was observed.

5. The apparatus according to claim 4, characterized in that, The first acquisition module includes: The first construction unit is used to determine multiple caries risk issues based on multiple preset caries risk factors, and to construct a caries risk questionnaire database through the multiple caries risk issues; The first determining unit is used to determine the risk weight of each option corresponding to each of the multiple caries risk issues, so as to obtain the risk score of each option through the risk weight of each option; The acquisition unit is used to acquire the caries risk data corresponding to the patient information to be trained based on the risk score of each option and the caries risk questionnaire database.

6. The apparatus according to claim 4, characterized in that, The training module includes: The second construction unit is used to construct the plurality of caries classification models, obtain each caries risk feature corresponding to the training dataset, and input each caries risk feature into each of the plurality of caries classification models to output the probability of all caries risk levels corresponding to each caries risk feature, wherein all caries risk levels include low-risk caries, medium-risk caries and high-risk caries; The weighted regression unit is used to perform a weighted regression operation on the probabilities of all caries risk levels output by the multiple caries classification models using the logistic regression algorithm, so as to determine the final caries risk level corresponding to each caries risk feature; The second determining unit is used to determine the weight of each caries risk feature based on each caries classification model; The weighted average unit is used to perform a weighted average operation on all weights corresponding to each caries risk feature to obtain the final weight of each caries risk feature; The sorting unit is used to sort the final weights of all caries risk features, generate the final weight sorting result, and determine the key risk features corresponding to the training dataset based on the final weight sorting result, so as to train the initial caries risk detection model through the key risk features.

7. A dental caries risk assessment device, used in the online prediction phase, characterized in that, The apparatus for assessing caries risk in the online prediction phase, as described in claim 3, comprises: The second data collection module is used to collect the personal information of the target patient and obtain the corresponding caries risk information of the target patient based on the preset caries risk questionnaire database. The preprocessing unit is used to perform data preprocessing operations on the patient's personal information and the caries risk information to generate data to be predicted; An evaluation unit is used to input the data to be predicted into a preset caries risk detection model to output caries risk detection results and generate a caries risk assessment report based on the caries risk detection results. The caries risk detection model is obtained by training a preset caries risk detection initial model with a preset training dataset.

8. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the method for assessing caries risk as described in any one of claims 1-2 or 3.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the caries risk assessment method as described in any one of claims 1-2 or 3.

Images