Oral health assessment method and system

Abstract

The invention provides an oral health assessment method and system, and belongs to the technical field of oral health management and artificial intelligence. The method comprises the following steps: step S110: obtaining inspection data of a target oral cavity; s120, determining an auxiliary database and an auxiliary recognition model based on the inspection data; s130, inputting the test data into an auxiliary identification model, and outputting an auxiliary diagnosis result; the inspection data of the target oral cavity obtained in the step S110 comprises oral cavity digital inspection image data and / or oral cavity microbiome sampling index data. The evaluation system comprises a test data acquisition unit, an auxiliary unit, a diagnosis unit and a retraining unit. According to the technical scheme, while oral health assessment based on the artificial intelligence auxiliary model can be realized, the defects of artificial intelligence diagnosis can be compared and displayed in teaching, and the database and the auxiliary model are continuously retrained in the diagnosis process, so that man-machine supplementation is realized to jointly improve the oral health assessment effect.

Description

Technical Field

[0001] The present invention belongs to the field of oral health management and artificial intelligence technology, and in particular relates to an oral health assessment method and system, a computer-readable storage medium for implementing the method, a computer program product, and an electronic device. Background Art

[0002] As an important organ of the human body, the health of the oral cavity is closely related to our physical condition. The oral health standard set by the World Health Organization is: the surface of the teeth is firm and clean, the color and shape of the gums are normal, there are no cavities, no obvious pain, and no bleeding. Oral abnormalities not only bring inconvenience to residents' physiological functions such as eating and speaking, but also cause diseases in other parts of the body such as diabetes, coronary heart disease, chronic gastritis, and even increase the risk of esophageal cancer, oral cancer, etc.

[0003] However, oral diseases are mostly chronic diseases. In the early stages, they are difficult to detect because they have no particularly obvious symptoms. Once obvious symptoms appear, such as gum pain and gum swelling, the condition is often serious. If you can conduct oral examinations on your own and detect oral abnormalities at an early stage, it may play a decisive role in preventing oral diseases.

[0004] The development and popularization of artificial intelligence (AI) technology has made it possible to implement the above ideas. The personalized oral health management system based on artificial intelligence technology starts from computer visualization technology and oral image analysis technology, explores the early oral symptom characteristics of different users, and uses the collected massive disease data for analysis and comparison to help users find early problems in oral health care. The oral disease and health management system based on artificial intelligence analysis proposed in the Chinese invention patent application with application number CN202311476060.6 provides a similar technical solution.

[0005] However, medical (disease) diagnosis is a highly professional discipline. The diagnosis of a specific disease must ultimately be based on the diagnosis of a professional doctor. The recommended results given by the AI ​​model can only be used as an intermediate result and cannot be provided to patients as the final result. However, due to the popularity of AI models, some oral patients, trainees (trainees, students, etc.) in oral-related disciplines, and even actual clinical staff overly rely on existing AI-assisted diagnosis technology, are unable to realize the defects of AI models, and lack their own subjective initiative, thereby reducing the accuracy of oral health assessments. Summary of the invention

[0006] In view of the above technical problems, the present invention proposes an oral health assessment method and system, a computer-readable storage medium, a computer program product and an electronic device for implementing the method.

[0007] In the first aspect of the present invention, an oral health assessment method is proposed, and the method is applied to the teaching process of oral health diagnosis.

[0008] The method includes the following steps:

[0009] S110: Obtain the test data of the target oral cavity;

[0010] S120: Determine the auxiliary database and the auxiliary recognition model based on the test data;

[0011] S130: Input the test data into the auxiliary recognition model and output the auxiliary diagnosis result;

[0012] Among them, the test data of the target oral cavity obtained in step S110 includes oral digital test image data and / or oral microbiome sampling index data.

[0013] The test data of the target oral cavity obtained in step S110 is anonymized oral test data;

[0014] Step S120 further includes:

[0015] Prompt the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral test data;

[0016] Based on the target age range and the test data, determine the auxiliary database from multiple candidate databases, and determine the auxiliary recognition model from multiple candidate recognition models. The multiple candidate recognition models include an index data self-learning diagnosis model and an image data recognition diagnosis model.

[0017] The multiple candidate databases include multiple oral health databases of different age ranges and different types.

[0018] The multiple candidate databases include an age range - microbiome sampling index database and an age range - oral digital test image database;

[0019] The age range - microbiome sampling index database stores data pairs of <microbiome sampling index, oral health assessment result>, and the age range - oral digital test image database stores data pairs of <test image feature value, oral health assessment result>;

[0020] After step S130, the method further includes:

[0021] When the difference degree between the auxiliary diagnosis result and the original diagnosis result of the target oral cavity is greater than the preset threshold, retrain the auxiliary recognition model using the test data;

[0022] The retraining includes:

[0023] Using the inspection data of the target oral cavity as an updated training sample, and using the inspection data of the auxiliary database corresponding to the auxiliary diagnosis result as an adversarial sample to retrain the auxiliary recognition model.

[0024] The specific steps of step S130 include:

[0025] Inputting the oral digital inspection image data into the image data recognition and diagnosis model, and the image data recognition and diagnosis model obtains a first auxiliary diagnosis result based on the age group - oral digital inspection image database;

[0026] And / or,

[0027] Inputting the oral microbiome sampling index data into the index data self - learning diagnosis model, and the index data self - learning diagnosis model obtains a second auxiliary diagnosis result based on the age group - microbiome sampling index database.

[0028] After obtaining the first auxiliary diagnosis result and / or the second auxiliary diagnosis result in step S130, the method includes the following steps:

[0029] S140: Obtaining the original diagnosis result of the target patient;

[0030] S150: Judging whether the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is normal based on the original diagnosis result of the target patient;

[0031] When the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is abnormal, updating the auxiliary database based on the inspection data of the target oral cavity, and retraining the auxiliary recognition model based on the updated auxiliary database.

[0032] The retraining of the auxiliary recognition model based on the updated auxiliary database specifically includes:

[0033] Using the inspection data of the target oral cavity as an updated training sample, and using the inspection data of the auxiliary database corresponding to the first auxiliary diagnosis result and / or the second auxiliary diagnosis result as an adversarial sample to retrain the auxiliary recognition model.

[0034] The foregoing oral health assessment method can be automatically implemented through computer program instructions on various forms of electronic devices; the computer program instructions can be stored in different forms of storage media and loaded into computer electronic devices for execution.

[0035] Therefore, in the second aspect of the present invention, there is also provided a computer-readable storage medium for storing computer instructions, which, when run on an electronic device, cause the electronic device to execute the foregoing oral health assessment method.

[0036] In the third aspect of the present invention, there is also proposed a computer device, which includes a processor and a memory. The memory is used to store instructions, and the processor is used to call the instructions in the memory, causing the computer device to execute the foregoing oral health assessment method.

[0037] In the fourth aspect of the present invention, there is also proposed a computer program product, which includes a computer program. When the computer program is executed, the foregoing oral health assessment method is implemented.

[0038] To implement the oral health assessment method in the foregoing first aspect, in the fifth aspect of the present invention, there is also proposed an oral health assessment system, which includes:

[0039] A test data acquisition unit for acquiring test data of a target oral cavity; the test data of the target oral cavity includes oral digital test image data and / or oral microbiome sampling index data;

[0040] An auxiliary unit for determining an auxiliary database and an auxiliary recognition model based on the test data;

[0041] A diagnosis unit for inputting the test data into the auxiliary recognition model and outputting an auxiliary diagnosis result;

[0042] A retraining unit for retraining the auxiliary recognition model with the test data when the difference between the auxiliary diagnosis result and the original diagnosis result of the target oral cavity is greater than a preset threshold;

[0043] The retraining includes:

[0044] Using the test data of the target oral cavity as updated training samples and the test data of the auxiliary database corresponding to the auxiliary diagnosis result as adversarial samples to retrain the auxiliary recognition model.

[0045] The system further includes:

[0046] An auxiliary diagnosis result sending unit for sending the auxiliary diagnosis result to the medical staff side when there is no original diagnosis result for the target oral cavity, and the medical staff side determines the health status of the target oral cavity based on the auxiliary diagnosis result.

[0047] The evaluation system is applied to the teaching process of oral health diagnosis, and the target oral examination data obtained by the examination data acquisition unit is anonymized oral examination data;

[0048] The auxiliary unit determines an auxiliary database and an auxiliary recognition model based on the examination data, specifically including:

[0049] Prompt the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral examination data;

[0050] Based on the target age range and the examination data, determine an auxiliary database from multiple candidate databases and determine an auxiliary recognition model from multiple candidate recognition models. The multiple candidate recognition models include an index data self-learning diagnosis model and an image data recognition diagnosis model.

[0051] The technical solution of the present invention can realize oral health assessment based on an artificial intelligence auxiliary model, and can contrast and display the defects of artificial intelligence diagnosis in teaching. During the diagnosis process, the database and the auxiliary model are continuously retrained, so that humans and machines complement each other to jointly improve the oral health assessment effect.

[0052] The further advantages of the present invention will be further detailed in the specific embodiment part in combination with the accompanying drawings of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0054] Figure 1 It is a schematic diagram of the main process of an oral health assessment method according to an embodiment of the present invention;

[0055] Figure 2 It is a schematic diagram of the main process of an oral health assessment method according to another preferred embodiment of the present invention;

[0056] Figure 3 It is a schematic diagram of the data flow control process involved in the oral health assessment process;

[0057] Figure 4 It is a schematic diagram of the functional module composition of an oral health assessment system according to an embodiment of the present invention;

[0058] Figure 5 It is a schematic diagram of the functional module composition of an oral health assessment system according to another preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] First, it should be pointed out that the embodiments of the oral health assessment method mentioned in this part can be implemented through a computer program on an electronic device or system configured with a memory and a processor. The forms of the electronic device or system can be a physical machine, a virtual machine, a server, a cluster, or any combination thereof.

[0060] Preferably, the specific form of the electronic device can also be a human-computer interaction terminal, and the human-computer interaction terminal can be a desktop terminal with a human-computer interaction interface, a smart handheld terminal, a mobile terminal, etc.

[0061] Figure 1 It is a schematic diagram of the main process of an oral health assessment method according to an embodiment of the present invention.

[0062] Figure 1 The method includes steps S110 - S130, and the specific implementation of each step is as follows:

[0063] S110: Obtain the test data of the target oral cavity;

[0064] S120: Determine the auxiliary database and the auxiliary recognition model based on the test data;

[0065] S130: Input the test data into the auxiliary recognition model and output the auxiliary diagnosis result.

[0066] When the specific embodiment Figure 1 is used for oral health assessment, the "auxiliary diagnosis result" should be understood as an "intermediate" result rather than the final result, because Figure 1 the embodiment can be understood as using computer-aided technology to evaluate the health status of the target oral cavity based on the existing test data to obtain an auxiliary health assessment result. However, those skilled in the art should understand that the final health assessment needs to be obtained by a doctor based on the above auxiliary diagnosis result and clinical experience.

[0067] However, due to the convenience and popularity of the AI model, some oral patients, trainees (students, etc.) in oral-related disciplines even actual clinical staff overly rely on the existing AI-assisted diagnosis technology, are unable to realize the defects of the AI model, lack subjective initiative, thus reducing the accuracy of oral health assessment.

[0068] Therefore, the first technical problem to be solved by the technical solution of this application is to show the defects of the AI model to relevant users (trainees (students, etc.) in oral-related disciplines, actual clinical staff) through a comparison method, and to enhance the subjective initiative of users during the AI-assisted diagnosis process.

[0069] At this time, as the first specific application and technical solution of the present invention, the method is applied to the teaching process of oral health diagnosis. The target oral test data obtained in step S110 is anonymized oral test data. The anonymization process at least includes eliminating data related to personal characteristics in the target oral test data, including name, age, gender, blood type, phone number, ID number, etc.

[0070] Specifically, on the basis of Figure 1 refer to Figure 2 , Figure 2 FIG. shows the main process schematic diagram of an oral health assessment method according to another preferred embodiment of the present invention. Figure 2 The preferred embodiment of Figure 1 After step S130 "output auxiliary diagnosis result" of the above embodiment, continue to execute the following steps S140 and S150:

[0071] S140: Obtain the original diagnosis result of the target patient;

[0072] S150: When it is determined that the auxiliary diagnosis result is abnormal based on the original diagnosis result of the target patient, retrain the auxiliary recognition model using the test data.

[0073] It can be seen that in this embodiment, the method is applied to the teaching diagnosis process. The target oral test data obtained in step S110 is anonymized oral test data. In fact, this oral test data comes from actual clinical practice. For example, a certain oral test data A0 of patient A, and a professional doctor has given a professional diagnosis conclusion B for patient A;

[0074] However, when the test data A0 is input into the auxiliary recognition model, the output auxiliary diagnosis result is A1;

[0075] When the difference degree between the auxiliary diagnosis result A1 and the original diagnosis result B of the target oral cavity is greater than a preset threshold, it is determined that the auxiliary diagnosis result is abnormal.

[0076] When specifically applying the method to the teaching process of oral disease diagnosis, the difference part between the auxiliary diagnosis result and the original diagnosis result can be contrastively displayed to the trainees (students, etc.) to be trained (in teaching) and actual clinical staff, so as to show the defects of the AI model and prompt the relevant personnel that the reliability of the diagnosis result of the AI model is insufficient.

[0077] Of course, at the same time, it is also necessary to remind the relevant personnel that although the AI model has deficiencies, it is itself a process of continuous learning and updating. Only through the complementarity of humans and machines can the oral health assessment effect be jointly improved.

[0078] Therefore, next, the auxiliary recognition model is retrained using the inspection data.

[0079] Specifically, the retraining includes:

[0080] Using the inspection data of the target oral cavity as updated training samples, and using the inspection data corresponding to the auxiliary diagnosis results in the auxiliary database as adversarial samples, the auxiliary recognition model is retrained.

[0081] Next, taking Figure 3 as an example, the data flow control process involved in the oral health assessment process is further introduced. It can be seen that in the subsequent embodiments, the oral health assessment method proposed in this application not only utilizes the AI model, but also interactively introduces user operations to select a matching database, avoiding the defects caused by over-reliance on AI technology.

[0082] Specifically, it can be seen from Figure 3 that the inspection data of the target oral cavity obtained in step S110 includes oral digital inspection image data and / or oral microbiome sampling index data.

[0083] As a specific example, the oral digital inspection image data includes oral imaging examination results, such as periapical films, panoramic films, cephalometric films, cone beam CT, etc.; it may also include other oral digital images and digital images.

[0084] Digital images capture and record every corner and detail of the oral cavity in the form of high-definition two-dimensional digital photos, precise digital impressions, and comprehensive three-dimensional surface scan data.

[0085] These images not only provide rich information about the patient's oral condition, but also provide a reliable visual basis for diagnosis and treatment planning; two-dimensional digital photos provide doctors with a quick way to examine the patient's current oral health condition with their intuitive and convenient characteristics; digital impressions transform the physical form of traditional impressions into digital information, greatly improving the efficiency of storage, replication, and analysis, while reducing the need for physical storage space; three-dimensional surface scan data shows the three-dimensional structure of the oral cavity in the form of a three-dimensional model, enabling diagnosis and treatment planning to achieve unprecedented precision.

[0086] Oral digital imaging technology is presented in a variety of forms, including high-resolution digital X-ray images, detailed cone beam CT (CBCT) images, comprehensive conventional CT images, MR I images with fine soft tissue presentation, and dynamically captured ultrasound images, etc.

[0087] The oral microbiome sampling index data is mainly reflected in the form of text and table-structured information. Oral bacterial diseases (such as dental caries and periodontal diseases) are closely related to the overgrowth of oral resident flora under pathogenic conditions, leading to oral microecological imbalance and opportunistic pathogenicity. Therefore, the relevant indicators of the bacterial genera significantly related to the severity of gingivitis can be used to diagnose the oral health index. Specifically, after oral bacterial sampling and pretreatment, an oral microbiome sampling index data report can be obtained.

[0088] The step S120 further includes:

[0089] prompting the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral examination data;

[0090] It can be seen that in step S120, the method needs to actively interact with the user, prompting the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral examination data.

[0091] The target age range can be a generalized age segment such as children, youth, middle-aged, or elderly, or it can be specific age segments such as 0-3 years old, 3-6 years old, 6-12 years old,... 18-30 years old... Of course, it can also be other age segment standards.

[0092] The reason for this is that there are significant differences in diet and living habits (such as smoking) among different age groups, resulting in different characteristic points of oral examination data. If a single database applicable to all age groups is uniformly used to perform subsequent AI model diagnosis, the accuracy is obviously low.

[0093] Therefore, multiple candidate databases can be pre-constructed, including multiple oral health databases for different age groups and different types.

[0094] Specifically, the multiple candidate databases include an age range - microbiome sampling index database and an age range - oral digital examination image database;

[0095] The age range - microbiome sampling index database stores data pairs of <microbiome sampling index, oral health assessment result>, and the age range - oral digital examination image database stores data pairs of <examination image eigenvalue, oral health assessment result>;

[0096] Therefore, to avoid the defects of AI model diagnosis, in step S120, it is necessary to give full play to the subjective initiative of professionals, prompting the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral examination data. In this process, the user can determine a target age range based on clinical experience and in combination with the relevant characteristic points of the oral examination data.

[0097] Furthermore, based on the target age group and the inspection data, an auxiliary database is determined from multiple candidate databases, and an auxiliary recognition model is determined from multiple candidate recognition models, where the multiple candidate recognition models include an index data self-learning diagnosis model and an image data recognition diagnosis model.

[0098] The step S130 specifically includes:

[0099] Input the oral digital inspection image data into the image data recognition diagnosis model, and the image data recognition diagnosis model obtains a first auxiliary diagnosis result based on the age group-oral digital inspection image database;

[0100] and / or,

[0101] Input the oral microbiome sampling index data into the index data self-learning diagnosis model, and the index data self-learning diagnosis model obtains a second auxiliary diagnosis result based on the age group-microbiome sampling index database.

[0102] The index data self-learning diagnosis model is a pre-trained index diagnosis model, and the basic principle of the diagnosis model is: determining the indexes of bacterial genera significantly related to the severity of oral (gingivitis), and constructing a microbiome-based oral (gingival) severity diagnosis model with this.

[0103] The image data recognition diagnosis model is also a pre-trained image diagnosis model, and the basic principle of the image diagnosis model is: based on technologies such as deep learning, neural networks, and data mining, the input oral digital inspection image data is preprocessed and then feature extraction, feature fusion, and classification are performed to obtain multiple features to be recognized, and based on the association relationship between the features to be recognized and oral abnormal indexes, the oral health status is determined.

[0104] It should be noted that whether it is the index diagnosis model or the image diagnosis model, they both belong to the mature existing technologies in this field, so they are not the key points of improvement of the present invention, and this embodiment does not expand on this specifically. However, for the AI diagnosis models (including the index diagnosis model and the image diagnosis model) mentioned in the existing technology, their training databases and interaction databases both uniformly use a universal database for the whole age group to perform subsequent AI model diagnosis, and their accuracy is obviously low; and there is no professional personnel assistance in the diagnosis process, covering up the defects of the AI model diagnosis results, and it is also not conducive to the interaction between AI and people.

[0105] The focus of the present invention is to introduce the interactive operation of professionals during the diagnosis process of the AI model, avoid the defects of the AI model-assisted diagnosis, so that the defects of the AI diagnosis can be compared and displayed in teaching, and continuously retrain the database and the auxiliary model during the diagnosis process, so that humans and machines complement each other to jointly improve the oral health assessment effect.

[0106] After obtaining the first auxiliary diagnosis result and / or the second auxiliary diagnosis result in the step S130, the method includes the following steps:

[0107] S140: Obtain the original diagnosis result of the target patient;

[0108] S150: Judge whether the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is normal based on the original diagnosis result of the target patient;

[0109] When the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is abnormal, update the auxiliary database based on the test data of the target oral cavity, and retrain the auxiliary recognition model based on the updated auxiliary database.

[0110] The retraining of the auxiliary recognition model based on the updated auxiliary database specifically includes:

[0111] Use the test data of the target oral cavity as the updated training sample, and use the test data of the auxiliary database corresponding to the first auxiliary diagnosis result and / or the second auxiliary diagnosis result as the adversarial sample to retrain the auxiliary recognition model.

[0112] After introducing the Figures 1 - 3 related details of the method embodiment of Figures 4 - 5 respectively give a schematic diagram of the functional module composition of an oral health assessment system in two different embodiments in a corresponding manner.

[0113] In Figure 4 corresponding to the method embodiment of Figure 2 a schematic diagram of an oral health assessment system is shown. The assessment system includes:

[0114] A test data acquisition unit for acquiring test data of a target oral cavity; the test data of the target oral cavity includes oral digital test image data and / or oral microbiome sampling index data;

[0115] An auxiliary unit for determining an auxiliary database and an auxiliary recognition model based on the test data;

[0116] A diagnosis unit for inputting the test data into the auxiliary recognition model and outputting an auxiliary diagnosis result;

[0117] A retraining unit for retraining the auxiliary recognition model using the test data when the difference between the auxiliary diagnosis result and the original diagnosis result of the target oral cavity is greater than a preset threshold;

[0118] The retraining includes:

[0119] Using the test data of the target oral cavity as updated training samples and the test data of the auxiliary database corresponding to the auxiliary diagnosis result as adversarial samples to retrain the auxiliary recognition model.

[0120] On Figure 4 this basis, Figure 5 the oral health assessment system shown further includes:

[0121] An auxiliary diagnosis result sending unit for sending the auxiliary diagnosis result to the medical staff side when there is no original diagnosis result for the target oral cavity, and the medical staff side judges the health status of the target oral cavity based on the auxiliary diagnosis result.

[0122] It should be understood that "the medical staff side judges the health status of the target oral cavity based on the auxiliary diagnosis result" here means that the "auxiliary diagnosis result" is only an "intermediate" result, rather than the final result. The final health assessment needs to be obtained by the doctor (medical staff side) based on the above auxiliary diagnosis result and clinical experience, and then finally judge the health status of the target oral cavity.

[0123] Preferably, when the system is specifically applied to the oral medicine teaching process, the target oral cavity test data obtained by the test data acquisition unit is anonymized oral cavity test data;

[0124] The auxiliary unit determines the auxiliary database and the auxiliary recognition model based on the test data, specifically including:

[0125] Prompting the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral cavity test data;

[0126] Based on the target age range and the test data, determining an auxiliary database from multiple candidate databases and determining an auxiliary recognition model from multiple candidate recognition models, where the multiple candidate recognition models include an index data self-learning diagnosis model and an image data recognition diagnosis model.

[0127] Preferably, although not shown, the system further includes a comparison display unit, configured to, when the difference degree between the auxiliary diagnosis result and the original diagnosis result of the target oral cavity is greater than a preset threshold, display the different parts between the auxiliary diagnosis result and the original diagnosis result to the personnel to be (taught) trained (trainees, students, etc.), and actual clinical staff, so as to display the defects of the AI model and prompt the relevant personnel that the reliability of the diagnosis result of the AI model is insufficient.

[0128] For other technologies, principles, algorithms or models not elaborated in detail in this application, reference can be made to the prior art.

[0129] Compared with the prior art, the prominent improvements and beneficial effects of the technical solution of this application at least include:

[0130] (1) The technical solution of the present invention can realize oral health assessment based on an artificial intelligence-assisted model, and can display the defects of artificial intelligence diagnosis for comparison in teaching.

[0131] (2) To avoid the defects of the AI model diagnosis, implementing the technical solution of the present invention requires giving play to the subjective initiative of professionals, and prompting the user to determine the target age range of the target patient corresponding to the target oral cavity according to the oral examination data. In this process, the user can determine a target age range based on clinical experience and in combination with relevant feature points of the oral examination data. For the multiple candidate databases constructed, including multiple oral health databases of different age ranges and different types, compared with the prior art where both the training database and the interaction database uniformly use a single general database for all age ranges to perform subsequent AI model diagnosis, the technical solution of this application can improve the accuracy;

[0132] (3) Continuously retraining the database and the auxiliary model during the diagnosis process, so that humans and machines complement each other to jointly improve the oral health assessment effect.

[0133] In the foregoing embodiment part, the present invention provides multiple embodiments, and each embodiment can constitute an independent technical solution and may contribute to the prior art and solve corresponding technical problems. However, it should be noted that different embodiments can be combined with each other without violating logic; at the same time, each embodiment can solve at least one technical problem, but it is not required that each individual embodiment solves multiple or all technical problems.

[0134] Meanwhile, in each specific implementation manner of this application, when it comes to relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first. Moreover, the collection, use, and processing of such data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of this application need to obtain the user's sensitive personal information, the user's separate permission or separate consent will be obtained through methods such as pop-up windows or redirecting to a confirmation page. After clearly obtaining the user's separate permission or separate consent, the necessary user-related data for the normal operation of the embodiments of this application will be obtained.

[0135] The various implementations of the present disclosure have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed implementations. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described implementations. The selection of the terms used herein is intended to best explain the principles of the implementations, the practical application, or the improvement of the technology in the market, or to enable other ordinary skill in the art in the technical field to understand the various implementation manners disclosed herein.

Claims

1. A method for evaluating oral health, characterized in that: The method comprises the following steps: S110: Acquire inspection data of the target oral cavity; S120: Determine an auxiliary database and an auxiliary recognition model based on the inspection data; S130: inputting the inspection data into the auxiliary recognition model, and outputting auxiliary diagnosis results; The target oral cavity inspection data acquired in step S110 includes oral cavity digital inspection image data and / or oral microbiome sampling index data.

2. An oral health assessment method according to claim 1, characterized in that: The method is applied to the oral health diagnosis teaching process, and the target oral examination data acquired in step S110 is oral examination data that has been anonymized; The step S120 further includes: Prompting the user to determine the target age group of the target patient corresponding to the target oral cavity according to the oral examination data; Based on the target age group and the inspection data, an auxiliary database is determined from multiple candidate databases, and an auxiliary recognition model is determined from multiple candidate recognition models, wherein the multiple candidate recognition models include an indicator data self-learning diagnosis model and an image data recognition diagnosis model.

3. An oral health assessment method as claimed in claim 2, characterized in that: The step S120 specifically includes: the multiple candidate databases include multiple oral health databases of different age groups and types.

4. An oral health assessment method according to claim 2 or 3, characterized in that: The multiple candidate databases include an age group-microbiome sampling index database and an age group-oral digital inspection image database; The age group-microbiome sampling index database stores <microbiome sampling index, oral health assessment result> data pairs, and the age group-oral digital inspection image database stores <inspection image feature value, oral health assessment result> data pairs; The step S130 specifically includes: Inputting the oral digital inspection image data into the image data recognition and diagnosis model, the image data recognition and diagnosis model obtains a first auxiliary diagnosis result based on the age group-oral digital inspection image database; and / or, The oral microbiome sampling index data is input into the index data self-learning diagnosis model, and the index data self-learning diagnosis model obtains a second auxiliary diagnosis result based on the age group-microbiome sampling index database.

5. An oral health assessment method as claimed in claim 4, characterized in that: After the first auxiliary diagnosis result and / or the second auxiliary diagnosis result are obtained in step S130, the method comprises the following steps: S140: Obtaining the original diagnosis result of the target patient; S150: Determine whether the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is normal based on the original diagnosis result of the target patient; When the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is abnormal, the auxiliary database is updated based on the inspection data of the target oral cavity, and the auxiliary recognition model is retrained based on the updated auxiliary database.

6. An oral health assessment method according to claim 5, characterized in that: The retraining of the auxiliary recognition model based on the updated auxiliary database specifically includes: The inspection data of the target oral cavity is used as an updated training sample, and the inspection data of the auxiliary database corresponding to the first auxiliary diagnosis result and / or the second auxiliary diagnosis result is used as an adversarial sample to retrain the auxiliary recognition model.

7. An oral health assessment system, characterized in that: The evaluation system comprises: A test data acquisition unit, used to acquire test data of a target oral cavity; the test data of the target oral cavity includes oral cavity digital test image data and / or oral microbiome sampling index data; an auxiliary unit, configured to determine an auxiliary database and an auxiliary recognition model based on the inspection data; A diagnosis unit, used for inputting the inspection data into the auxiliary recognition model and outputting an auxiliary diagnosis result; a retraining unit, configured to retrain the auxiliary recognition model using the inspection data when the difference between the auxiliary diagnosis result and the original diagnosis result of the target oral cavity is greater than a preset threshold; The retraining includes: The inspection data of the target oral cavity is used as an updated training sample, and the inspection data of the auxiliary database corresponding to the auxiliary diagnosis result is used as an adversarial sample to retrain the auxiliary recognition model.

8. An oral health assessment system as claimed in claim 7, characterized in that: The system further comprises: The auxiliary diagnosis result sending unit is used to send the auxiliary diagnosis result to the medical care end when there is no original diagnosis result for the target oral cavity, and the medical care end determines the health status of the target oral cavity based on the auxiliary diagnosis result.

9. An oral health assessment system as claimed in claim 7, characterized in that: The evaluation system is applied to the oral health diagnosis teaching process, and the target oral examination data acquired by the examination data acquisition unit is oral examination data that has been anonymized; The auxiliary unit determines an auxiliary database and an auxiliary recognition model based on the inspection data, specifically including: Prompting the user to determine the target age group of the target patient corresponding to the target oral cavity according to the oral examination data; Based on the target age group and the inspection data, an auxiliary database is determined from multiple candidate databases, and an auxiliary recognition model is determined from multiple candidate recognition models, wherein the multiple candidate recognition models include an indicator data self-learning diagnosis model and an image data recognition diagnosis model.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by a processor to implement the oral health assessment method according to any one of claims 1 to 6.

Images