KR20210110214A - Prediction systems or methods for possible dental treatment plan - Google Patents

Abstract

Disclosed, in the present invention, are a system and method for predicting a dental treatment, wherein the system comprises: a data acquisition part that acquires treatment data of a plurality of patients undergoing dental treatment; a prediction model generating part that generates a prediction model that learns a correlation between the treatment data and disease occurrence based on the plurality of acquired treatment data; a data input part that receives data of a first visit before dental treatment from a specific patient; and a prediction result output part that outputs the prediction result data by applying the inputted state data to the prediction model.

Description

DENTAL TREATMENT PREDICTION SYSTEMS AND METHODS

The present invention relates to a dental treatment prediction system and method.

Treatment methods for a patient's dental disease (cavity, tooth fracture, tooth defect, periodontitis, etc.) include tooth extraction, nerve treatment, crown, bridge, implant, and denture treatment.

In this case, rather than just one treatment, multiple treatment methods are usually suggested based on the patient's medical history (basic disease such as diabetes or hypertension) and dental history (existing dental treatment), and the And it is common to select what is judged to be the optimal dental treatment by dental judgment.

However, in the case of non-dental specialists, the presentation of dental treatment methods that can be suggested in the future is limited depending on the current patient situation, and even if it is a specialist, the future treatment plans presented by many specialists do not match, and the In this case, as it is difficult to predict the scope of treatment and cost of treatment in the future, consultation through individual visits to multiple dental clinics is used.

In addition, as it is difficult to estimate the scope of treatment and cost of treatment in the future even in the case of insurance company premium calculation officers, only a simple classification method using very limited data such as gender and age is presented for the current dental insurance premium calculation.

An object of the present invention is to provide a dental treatment prediction system and method capable of predicting future dental treatment based on various dental related data provided as text and image data of a patient with dental disease.

On the other hand, the technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned are clearly to those of ordinary skill in the art to which the present invention belongs from the description below. can be understood

A dental treatment prediction system according to an embodiment of the present invention includes: a data acquisition unit for acquiring treatment data of a plurality of patients who have undergone dental treatment; a predictive model generator for generating a predictive model that learns a correlation between treatment data and disease occurrence based on the plurality of obtained treatment data; a data input unit for receiving state data from a specific patient; and a prediction result output unit for outputting prediction result data by applying the input state data to the prediction model. may include.

In addition, the data acquisition unit may normalize the obtained treatment data data.

In addition, the treatment data may include pre-treatment data and post-treatment data of the patient, and each of the pre-treatment data and post-treatment data may include text data and image data.

In addition, the predictive model may be trained by using the pre-treatment data as input data and the post-treatment data as output data.

In addition, the predictive model learns the correlation between the treatment data and the disease occurrence and the treatment cost of the disease, and when the state data of a specific patient is input, the probability of occurrence of the disease of the specific patient and the treatment cost of the possible diseases predictable.

In addition, further comprising a communication unit capable of transmitting and receiving data, wherein the communication unit receives treatment data and status data from a patient terminal, an expert terminal, or a management server, and transmits the output prediction result data to a patient terminal, an expert terminal, a management server, or an insurance company can be sent to the server.

In addition, a dental treatment prediction method according to an embodiment of the present invention includes: a data acquisition step of acquiring treatment data of a plurality of patients who have undergone dental treatment; a predictive model generation step of generating a predictive model that learns a correlation between treatment data and disease occurrence based on the plurality of obtained treatment data; a data input step of receiving state data from a specific patient; and a prediction result output step of outputting prediction result data by applying the input state data to the prediction model. may include.

In addition, the data acquisition step may be data normalization of the acquired treatment data.

In addition, the treatment data acquired in the data acquisition step may include pre-treatment data and post-treatment data of the patient, and each of the pre-treatment data and post-treatment data may include text data and image data.

In addition, in the predictive model generated in the predictive model generating step, the pre-treatment data may be used as input data and the post-treatment data may be used as output data to perform learning.

In addition, the predictive model learns the correlation between the treatment data and the disease occurrence and the treatment cost of the disease, and when the state data of a specific patient is input, the probability of occurrence of the disease of the specific patient and the treatment cost of the possible diseases predictable.

According to an embodiment of the present invention, it is possible to predict future dental treatment based on various dental related data provided as text and image data of a patient with dental disease.

On the other hand, the effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be able

1 is a block diagram schematically showing a medical treatment system according to an embodiment of the present invention;
2 is a block diagram schematically showing the configuration of a dental treatment prediction system according to an embodiment of the present invention;
3 is a flowchart illustrating a method for predicting dental treatment according to an embodiment of the present invention;
4 and 5 are exemplary views showing data used for learning a predictive model according to the present invention,
6 is an exemplary diagram illustrating a learning process of a predictive model according to the present invention;
7 is an exemplary diagram of a process of dataizing the pre-treatment data input for learning in FIG. 6;
8 is an exemplary diagram of a process of converting data after treatment input for learning in FIG. 6 ,
9 is an exemplary view showing a prediction result through a prediction model according to the present invention,
10 is an exemplary view showing an example of applying the dental treatment prediction system according to the present invention to insurance design,
11 is an exemplary diagram illustrating an embodiment of an artificial intelligence-based non-face-to-face and automated dental insurance eligibility review process according to the present invention.

Hereinafter, an embodiment of the present invention will be described in more detail with reference to the accompanying drawings. Embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited to the following embodiments. This embodiment is provided to more completely explain the present invention to those of ordinary skill in the art. Accordingly, the shapes of elements in the drawings are exaggerated to emphasize a clearer description.

The configuration of the invention for clarifying the solution of the problem to be solved by the present invention will be described in detail with reference to the accompanying drawings based on a preferred embodiment of the present invention, but the same in assigning reference numbers to the components of the drawings For the components, even if they are on different drawings, the same reference numbers are given, and it is noted in advance that the components of other drawings can be cited when necessary in the description of the drawings.

1 is a block diagram schematically illustrating a medical treatment system according to an embodiment of the present invention.

Referring to FIG. 1 , a medical treatment system according to an embodiment of the present invention may include a terminal 10 , a management server 20 , an insurance company server 30 , and a dental treatment prediction system 100 .

The terminal 10, the management server 20, the insurance company server 30, and the dental treatment prediction system 100 may be interconnected through the communication network 1, and may mutually transmit/receive dental-related treatment data and prediction result data. .

The terminal 10 may include a user terminal 11 and an expert terminal 12 .

The user terminal 11 may be a mobile device carried by a patient, and may be configured as a type of PC, smart phone, pad, wearable device, notebook computer, or PDA.

Hereinafter, the user may be used interchangeably with the patient.

In addition, the expert terminal 12 may be a mobile device possessed by a dental professional such as a doctor, and may be composed of a PC, a smartphone, a pad, a wearable device, a notebook computer, or a PDA.

Here, the user terminal 11 may store its own dental treatment data, and the expert terminal 12 may store treatment data of a plurality of patients treated by the user terminal 11 .

Meanwhile, an application related to the medical treatment system according to an embodiment of the present invention may be stored and driven in each of the user terminal 11 and the expert terminal 12 .

The management server 20 may operate the overall treatment system, and may store treatment data of a plurality of patients received from the user terminal 11 and the expert terminal 12 . In addition, the management server 20 may store the prediction result of a specific patient output from the dental treatment prediction system 100 , and transmit it to the user terminal 11 , the expert terminal 12 , and the insurance company server 30 .

The insurance company server 30 may predict the insurance premium payment range based on the prediction result of a specific patient, and may evaluate the payability through this.

Figure 2 is a block diagram schematically showing the configuration of a dental treatment prediction system according to an embodiment of the present invention, Figure 3 is a flowchart showing a dental treatment prediction method according to an embodiment of the present invention, Figures 4 and 5 is an exemplary diagram illustrating data used for learning a predictive model according to the present invention, FIG. 6 is an exemplary diagram illustrating a learning process of a predictive model according to the present invention, and FIG. 7 is a prediction result through a predictive model according to the present invention is an exemplary view, and FIG. 8 is an exemplary view showing an example of applying the dental treatment prediction system according to the present invention to insurance design.

Hereinafter, the dental treatment prediction system 100 and a dental treatment prediction method through the same will be described with reference to FIGS. 2 to 8 together.

First, referring to FIG. 2 , the dental treatment prediction system 100 includes a communication unit 110 , a data acquisition unit 120 , a prediction model generation unit 130 , a data input unit 140 , and a prediction result output unit 150 . may include

The communication unit 110 performs a communication function to exchange data between the communication network 1 and the dental treatment prediction system 100 .

The communication unit 110 includes Bluetooth communication, BLE (Bluetooth Low Energy) communication, near field communication unit, WLAN (Wi-Fi) communication, Zigbee communication, infrared (IrDA, infrared Data Association) communication, Short-distance communication such as Wi-Fi Direct (WFD) communication, ultra-wideband (UWB) communication, and Ant+ communication may be provided. According to an embodiment, the communication unit 160 may provide a mobile communication function such as 2G, 3G, 4G, or WiBro.

The data acquisition unit 120 may acquire treatment data of a plurality of users from the user terminal 11 , the expert terminal 12 , and the management server 20 through the communication unit 110 ( S10 ).

Here, each of the user's treatment data may include text data and image data.

Here, as shown in FIG. 4, the text data includes a medical certificate used by an insurance company or an electronic medical record book and a handwritten medical record used in a dental hospital, such as the patient's age, gender, underlying disease, and surgery history. Initial dental records such as medical history, missing teeth, periodontal disease teeth, carious teeth, and prosthetic treatment data, and individual dental treatment records and treatment costs after completion of dental treatment such as extraction, implant, bridge, denture, filling treatment, etc. It may include medical record data in the form of text related to the patient.

In addition, as shown in FIG. 5 , the image data may include medical image data related to the patient, such as dental panoramic X-rays, intraoral photographs, CBCT, and digital oral scan data.

As described above, the data acquisition unit 120 can check the patient's general condition and dental condition before and after treatment, such as initial medical records and dental treatment confirmations for a plurality of patients owned by hospitals, clinics, insurance companies, etc. It is possible to build and collect data in the form of images that can check the condition before and after treatment with text-type data that can be used and intraoral photos and dental X-ray data as a structured digital database.

Here, the data acquisition unit 120 may convert the medical record data and image data in the form of text into digital data, and may additionally perform various data pre-processing processes such as normalization.

The predictive model generating unit 130 generates a predictive model that learns the correlation between the treatment data and the disease occurrence based on the data obtained by the data obtaining unit 120 ( S20 ).

The predictive model generator 130 applies various artificial intelligence algorithm technologies to learn the pre-treatment state based on digital form data, and then when any pre-treatment state data is input, the dental disease progression trend and possible dental treatment methods and an artificial intelligence model that predicts cost ranges.

Here, the predictive model generator 130 may generate a model for predicting a learning algorithm based on artificial intelligence based on the correlation between the initial dental treatment data and the corresponding dental treatment occurrence and treatment cost. The AI-based learning algorithm models that can be applied at this time are machine learning-based algorithm models, such as Multilayer perceptron, Na ve-Bayesian classification, and Random forest classification. etc., Convolutional Neural Network, Generative Adversarial Networks, etc. can be used as algorithm models based on Deep Learning. In addition, various data augmentation techniques and reinforcement learning techniques are currently available It can be broadly applied to all predictable AI learning-based algorithms that have been or will be developed in the future.

On the other hand, as software for performing machine learning, Weka (Waikato Environment for Knowledge Analysis), a free software licensed under the GNU General Public License, and a programming language that is an open, community-based development model managed by the non-profit Python Software Foundation. Python, R, a programming language and open source software for statistical calculations and graphics, and MATLAB, an engineering software that provides numerical analysis and programming environment developed by MathWorks for a fee, has been released or in the future. All foreseeable artificial intelligence programming languages and platforms being developed can be broadly applied.

In the present invention, a predictive model can be designed for each type of machine learning, and further, a model can be designed using an analysis method of deep learning corresponding to a high-dimensional analysis of machine learning.

That is, the predictive model generator 130 may input the data acquired by the data acquisition unit 120 by connecting the multilayer perceptron neural network to the cyclic neural network. In addition, the predictive model generator 130 may sequentially input and analyze the correlation between the variables as well as the correlation over time of each dynamic variable through a plurality of repeatedly measured state variables, by sequentially inputting them into the cyclic neural network. have.

Meanwhile, referring to FIG. 6 , in an embodiment, the artificial intelligence prediction model may be defined in the form 131 of an artificial intelligence neural network, and text data and images for each individual patient having data before and after treatment. It can be generated through artificial intelligence algorithm-based learning by using the pre-treatment state as the input data 132 and the post-treatment state as the output data 133 among the data.

6 and 7, the data used as input data 132 for learning in the predictive model generating unit 130 is a plurality of individual patient-specific data, and the patient's gender, age, and whole body on a medical certificate or medical record sheet. Medical history including diseases and text data of each tooth condition (defect, caries, periodontal disease, prosthetic treatment, restorative treatment, etc.) are converted into digital data for each item and used as input variables. , X-ray photos, etc. are used as input variables in the form of multi-dimensional digital numerical data through a multi-dimensional vector digitization change operation through a feature extraction technique as input data for training the artificial intelligence neural network 131. can be used, and examples of variables of input data in FIG. 7 may be further diversified according to the types of inputted medical confirmation and medical records.

6 and 8 , the data used as the output data 133 for learning means data of the state after dental treatment corresponding to the input data of the state before dental treatment of the individual patient, and the input data Text data for dental treatment records such as tooth extraction, implant treatment, crown treatment, bridge treatment, restorative treatment, denture treatment, etc., and image data such as oral photos and X-ray photos after dental treatment on the medical records used in the data are artificial intelligence It can be used as output data for training the neural network 131, and examples of variables of data input as the output data 133 in FIG. .

The data input unit 140 may input current state data for a specific patient into the predictive model ( S30 ).

Here, the state data input through the data input unit 140 is text data (recording paper in the form of a text that can know the patient's first visit status, such as a dental first visit record sheet) and image data (a patient's intraoral photo and dental treatment) related to the pre-treatment state. X-ray data related to) may include one or more of.

The prediction result output unit 150 may output result data corresponding to the state data input to the prediction model ( S40 ).

Referring to Figure 9, the first visit data (gender, age, medical history, including systemic disease, etc.) of any patient who is a prediction target using a predictive model, and each tooth situation (defect, tooth decay, periodontal disease, Text data of prosthetic treatment, restorative treatment, etc.) and image data such as oral photos and X-ray photos before dental treatment are used as input data 134 to the artificial intelligence neural network 131, and the artificial intelligence neural network 131 is a probability Dental treatment methods with a high probability of occurrence may be output as text (expected tooth extraction, implant treatment, crown treatment, bridge treatment, restorative treatment, denture treatment, etc.) and output data 135 in the form of an image.

Meanwhile, data output from the prediction result output unit 150 may be used as basic data for calculating a dental treatment cost range and dental insurance premium. Here, referring to FIG. 8 as an example, when a random patient inputs (uploads) one or more first visit records in the form of text or images to the system before dental treatment through the user terminal 11 in order to sign up for dental insurance, The dental treatment prediction system 100 may predict a dental treatment range with a high probability of occurrence using a predictive model, and use it as basic data for predicting a range of dental treatment costs and calculating insurance premiums based on this.

Also, referring to FIG. 10 , as described above, the data output from the prediction result output unit 150 may be provided to the management server 20 or the insurance company server 30 through the communication unit 110 , and the insurance company Based on the output result, the server 30 may predict a range of dental treatment costs according to the possibility of future occurrence of a specific patient, and may calculate an insurance premium for a specific patient based on this.

In addition, referring to FIG. 11 , a person who intends to purchase dental insurance transmits the patient's first examination record or dental X-ray to a digital device such as a mobile device after performing the initial examination at any dental office to a program such as an insurance application At the time, the learned artificial intelligence system extracts information about the patient's first visit status from the patient's first visit record sheet or dental X-ray. Information on the patient's first visit status includes information such as the patient's current number of remaining teeth, the number according to the type and type of dental prosthesis restoration, the current tooth defect status and the number, and the number of teeth in the periodontitis state with a high probability of tooth extraction. characteristics of the patient's teeth.

The artificial intelligence system, which learned the initial dental examination status data in advance and the treatment information and insurance premium claim data after dental treatment occurred to the insurance company, is Based on the data, it is possible to predict the type of dental treatment that will occur in the future and the range of the insurance premium billing amount, etc., and provide the information to the insurance company or related parties for use.

The above detailed description is illustrative of the present invention. In addition, the above description shows and describes preferred embodiments of the present invention, and the present invention can be used in various other combinations, modifications, and environments. That is, changes or modifications are possible within the scope of the concept of the invention disclosed herein, the scope equivalent to the written disclosure, and/or within the scope of skill or knowledge in the art. The written embodiment describes the best state for implementing the technical idea of the present invention, and various changes required in the specific application field and use of the present invention are possible. Accordingly, the detailed description of the present invention is not intended to limit the present invention to the disclosed embodiments. Also, the appended claims should be construed as including other embodiments.

10: terminal
20: management server
30: Insurance company server
100: Dental Care Prediction System

Claims (11)

a data acquisition unit configured to acquire treatment data of a plurality of patients undergoing dental treatment;
a predictive model generator for generating a predictive model that learns a correlation between treatment data and disease occurrence based on the plurality of obtained treatment data;
a data input unit for receiving state data from a specific patient; and
a prediction result output unit that applies the input state data to the prediction model and outputs prediction result data; A dental treatment prediction system comprising a.
The method of claim 1,
The data acquisition unit is a dental treatment prediction system for data normalizing the acquired treatment data.
3. The method of claim 2,
the treatment data includes pre-treatment data and post-treatment data of the patient;
Each of the pre-treatment data and the post-treatment data includes text data and image data.
4. The method of claim 3,
The predictive model is
The pre-treatment data is used as input data,
A dental treatment prediction system for learning by using the post-treatment data as output data.
5. The method of claim 4,
The predictive model is
learning the correlation between the treatment data and the occurrence of the disease and the cost of treatment of the disease,
When the state data of a specific patient is input, a dental treatment prediction system for predicting the disease occurrence probability of the specific patient and the treatment cost of the possible diseases.
6. The method of claim 5,
Further comprising a communication unit capable of transmitting and receiving data,
The communication unit,
receiving treatment data and status data from a patient terminal, an expert terminal, or a management server;
A dental treatment prediction system that transmits the output prediction result data to a patient terminal, an expert terminal, a management server, or an insurance company server.
A data acquisition step of acquiring treatment data of a plurality of patients who have undergone dental treatment;
a predictive model generation step of generating a predictive model that learns a correlation between treatment data and disease occurrence based on the plurality of obtained treatment data;
a data input step of receiving state data from a specific patient; and
a prediction result output step of outputting prediction result data by applying the input state data to the prediction model; A method of predicting dental care comprising a.
8. The method of claim 7,
The data acquisition step is
A dental treatment prediction method for data normalizing the obtained treatment data.
9. The method of claim 8,
In the data acquisition step
The obtained treatment data includes pre-treatment data and post-treatment data of the patient,
The pre-treatment data and the post-treatment data each include text data and image data.
10. The method of claim 9,
The predictive model generated in the predictive model generation step is,
The pre-treatment data is used as input data,
A dental treatment prediction method for learning by using the post-treatment data as output data.
11. The method of claim 10,
The predictive model is
learning the correlation between the treatment data and the occurrence of the disease and the cost of treatment of the disease,
When the state data of a specific patient is input, a dental treatment prediction method for predicting the probability of occurrence of a disease of the specific patient and the cost of treatment of possible diseases.

Images