KR20220164149A - Method for predicting depression occurrence using artificial intelligence model and computer readable record medium thereof - Google Patents

Abstract

The present invention relates to a method for predicting the depression occurrence using an artificial intelligence model. The method of the present invention may include: a data receiving step in which a processor of a computing device receives big data including biological data, psychological data, and sociological data from a server connected to the device; a first parameter extraction step in which a processor extracts a parameter related to the occurrence of depression from the received data; and a depression occurrence prediction step of predicting the depression occurrence through a result value obtained by inputting the extracted first parameter into the learned artificial intelligence model.

Description

Method for predicting depression occurrence using artificial intelligence model and computer readable recording medium recording it {Method for predicting depression occurrence using artificial intelligence model and computer readable record medium thereof}

The present invention relates to a method for predicting the onset of depression using an artificial intelligence model and a computer-readable recording medium recording the same, and more particularly, to extract parameters from clinical research big data, and use the extracted parameters to predict the onset of depression using an artificial intelligence model It relates to a method for predicting the occurrence of depression using an artificial intelligence model that predicts and a computer-readable recording medium recording the same.

As a recent global economic and social environmental problem, depression is a universal social phenomenon or symbol of the present age, beyond a simple individual mental illness, due to low growth and high unemployment, deepening inequality in various dimensions, and consequent changes and dissolution of family structures. as a social problem.

Therefore, there is a need for research on the resolution and management of depression at the social and national level beyond the individual pathological level, and people with specific biological, psychological, and social characteristics are There is a need for a system that can grasp the social context of how this process of change leads to entering the high-risk group of depression and the mechanism of depression at a more diverse level.

Therefore, if depression can be predicted and prevented at an early stage by establishing a system that can deliver, accumulate, and predict individual information that is difficult to solve face-to-face through the recently developed ICT technology, it is very important in academic, clinical, and social problem solving. can be a useful alternative.

[Patent Document 1] Korean Patent Registration No. 10-2111374. 2020.05.11. registration.

The present invention has been created to solve the above problems, and an object of the present invention is to provide a method for predicting the occurrence of depression using an artificial intelligence model and a computer readable recording medium recording the same.

The objects of the present invention are not limited to the objects mentioned above, and other objects not mentioned will be clearly understood from the description below.

In order to achieve the above objects, a method for predicting the occurrence of depression using an artificial intelligence model according to an embodiment of the present invention and a computer readable recording medium recording the same are disclosed. The method includes a data receiving step in which a processor of a computing device receives big data including biological data, psychological data, and sociological data from a server connected to the device, and extracting parameters related to the occurrence of depression among the data received by the processor. It may include a first parameter extraction step, a depression occurrence prediction step of predicting the occurrence of depression through a result value obtained by inputting the extracted first parameter to the learned artificial intelligence model.

In addition, according to an embodiment of the present invention, big data includes bio-signal data measured through sensors, psychological test data through surveys, social network survey data through surveys, and activity data using communication networks. can be characterized.

In addition, according to an embodiment of the present invention, the step of predicting the occurrence of depression is performed by determining the first parameter as the contribution of the occurrence of depression through decision tree analysis and regression analysis on the extracted first parameter. A second parameter extraction step of digitizing and extracting a parameter with a high contribution among the digitized first parameters and predicting the occurrence of depression through the result value obtained by inputting the extracted second parameter to the artificial intelligence model steps may be included.

In addition, according to an embodiment of the present invention, the artificial intelligence model may be characterized by using an ensemble learning (EL) method as a prediction algorithm.

In addition, according to an embodiment of the present invention, the artificial intelligence model may be characterized in that the second parameter is determined as a final value by using a plurality of machine learning algorithms and an overall average value of the respective result values.

In addition, according to one embodiment of the present invention, it may be a computer-readable recording medium on which a program for executing a method for predicting the occurrence of depression using an artificial intelligence model in a computer is recorded.

Specific details for achieving the above objects will become clear with reference to embodiments to be described later in detail in conjunction with the accompanying drawings.

However, the present invention is not limited to the embodiments disclosed below, and may be configured in a variety of different forms, so that the disclosure of the present invention is complete and those of ordinary skill in the art to which the present invention belongs ( It is provided hereafter to fully inform the "ordinary skilled person") of the scope of the invention.

According to an embodiment of the present invention, it is possible to increase the predictability of depression using an artificial intelligence model.

In addition, according to an embodiment of the present invention, a user can easily predict the occurrence of depression through a computing device without the help of an expert.

The effects of the present invention are not limited to the above-mentioned effects, and the potential effects expected by the technical features of the present invention will be clearly understood from the description below.

In order that the above-mentioned features of the present invention may be understood in detail, more specifically, with reference to the following embodiments, some of the embodiments are shown in the accompanying drawings. Also, like reference numbers in the drawings are intended to refer to the same or similar function throughout the various aspects. However, it should be noted that the accompanying drawings only show specific exemplary embodiments of the present invention, and are not considered to limit the scope of the present invention, and other embodiments having the same effect may be fully appreciated. let it do
1 is a diagram showing a block diagram of a depression occurrence prediction system using an artificial intelligence model according to an embodiment of the present invention.
2 is a flowchart of a method for predicting the occurrence of depression using an artificial intelligence model according to an embodiment of the present invention.
Figure 3 is a flow chart of the step of predicting the occurrence of depression according to an embodiment of the present invention.
4 is a diagram illustrating parameters extracted according to the degree of contribution to occurrence of depression according to an embodiment of the present invention.
5 is a diagram illustrating an ensemble learning model according to an embodiment of the present invention.
6 is a diagram showing a result of predicting the occurrence of depression using an artificial intelligence model according to an embodiment of the present invention.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail.

Various features of the invention disclosed in the claims may be better understood in consideration of the drawings and detailed description. Devices, methods, manufacturing methods, and various embodiments disclosed in the specification are provided for illustrative purposes. The disclosed structural and functional features are intended to enable a person skilled in the art to specifically implement various embodiments, and are not intended to limit the scope of the invention. The disclosed terms and phrases are intended to provide an easy-to-understand description of the various features of the disclosed invention, and are not intended to limit the scope of the invention.

In describing the present invention, if it is determined that a detailed description of related known technologies may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

Hereinafter, a method for predicting depression using an artificial intelligence model according to an embodiment of the present invention will be described.

1 is a diagram showing a block diagram of a depression occurrence prediction system using an artificial intelligence model according to an embodiment of the present invention.

Referring to FIG. 1 , a depression occurrence prediction system using an artificial intelligence model may include a processor 100, a memory unit 200, and a network unit 300.

In one embodiment, the processor 100 is a device that controls the entire computer system, and can control and adjust a series of processes that receive data from various input devices, process them, and send the results to an output device. .

In one embodiment, the processor 100 according to the present disclosure may perform the functions of the electronic device illustrated in FIG. 1 by executing programs stored in the memory 200 . In addition, the processor 100 may be composed of one or a plurality of processors, and the one or plurality of processors may be a general-purpose processor such as a CPU, AP, DSP (Digital Signal Processor), a graphics-only processor such as a GPU, or artificial intelligence (AI). ) may be a dedicated processor. According to an embodiment, when the processor 100 includes a general-purpose processor, an artificial intelligence processor, and a graphics-only processor, the artificial intelligence processor may be implemented as a separate chip from the general-purpose processor or the graphics-only processor.

According to an embodiment, when the processor 100 is implemented as a plurality of processors, graphics-only processors, or artificial intelligence-only processors, at least some of the plurality of processors, graphics-only processors, or artificial intelligence-only processors are connected to electronic devices and electronic devices. It may be mounted on another connected electronic device or server.

According to an embodiment, the computing device may include an artificial neural network and process weights of the artificial neural network. For example, the computing device may quantize the weights of the artificial neural network and encode at least some of the quantized weights.

According to an embodiment, the computing device includes a smart phone, tablet PC, PC, smart TV, mobile phone, media player, server, micro server, It may be other mobile or non-mobile computing devices, but is not limited thereto.

Also, the processor 100 may receive big data including biological data, psychological data, and sociological data from a server connected to the device.

Here, big data may include bio-signal data measured through sensors, psychological test data through surveys, social network survey data through surveys, and activity data using communication networks.

More specifically, big data may include clinical research data, survey data responding to preset questions and answers, psychological tests, user biometric data measured through sensors of electronic devices, and SNS activities used by users through communication devices. , activity data such as text messages or phone calls, and the like. The present invention can increase predictability by using big data information that is a synthesis of biometric data that can be measured in real time and objectively confirmed numerically based on sociological and psychological data and used as predictive data for the occurrence of depression, , can guarantee the accuracy of prediction results.

In addition, biometric data may include biosignals such as real-time heart rate, amount of exercise, amount of sleep, etc. Psychological test data is data to respond to short questions about mood, mood, and mental health status periodically or non-periodically. Network survey data may include information such as characteristics of social relationships (eg, size of personal network, type of personal network, strength of personal network, etc.), amount of phone calls, amount of social activity using GPS, etc., and SNS activity data.

Also, the processor 100 may extract a parameter related to occurrence of depression from among the received data.

More specifically, parameters including factors directly or indirectly related to the occurrence of depression may be primarily extracted from the big data received from the server.

In addition, the processor 100 may predict occurrence of depression through a result obtained by inputting the extracted first parameter to the learned artificial intelligence model.

More specifically, the processor 100 quantifies the first parameter according to the contribution to the occurrence of depression through decision tree analysis and regression analysis on the extracted first parameter, and quantifies the digitized first parameter. A second parameter for extracting a parameter with a high contribution among the first parameters may be extracted, and the occurrence of depression may be predicted through a result value obtained by inputting the extracted second parameter to an artificial intelligence model.

In addition, weights may be assigned to parameters having a high contribution to the occurrence of depression, and the prediction of occurrence of depression may increase as parameters having a high contribution are included.

Here, the artificial intelligence model, as a prediction algorithm, may include an ensemble learning (EL) method. In addition, the artificial intelligence model may use a plurality of machine learning algorithms for the second parameter having a high contribution to the occurrence of depression, and determine the overall average value of the respective result values as the final value. The accuracy of prediction can be increased by using the above method.

In addition, the artificial intelligence model may be characterized by learning information related to the occurrence of depression using the aforementioned big data.

In one embodiment, the memory unit 200 may store programs for processing and control of the processor 100 and may store data input to or output from the electronic device. In addition, the memory unit 200 includes information on weights related to layers constituting the artificial neural network, nodes included in the layers, and connection strengths of the layers, a weight normalization function, a quantization function, and types and identifications of the identified weights. Encoded weight values may be stored according to the type of the weight. That is, when the weights in the artificial neural network are modified and updated, the memory 200 may further store information about the modified and updated weights.

In addition, the memory unit 200 may be a flash memory type, a hard disk type, a multimedia card micro type, or a card type memory (for example, SD or XD memory, etc.) ), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic It may include at least one type of storage medium among a memory, a magnetic disk, and an optical disk, but the present disclosure is not limited thereto.

Also, the memory unit 200 may extract and store parameters related to the occurrence of depression among the received data. In addition, the memory unit 200 may store a result of predicting occurrence of depression through a result obtained by inputting the first parameter extracted by the processor 100 to the learned artificial intelligence model.

In one embodiment, the network unit 300 may enable a computing device to exchange information with a user terminal through wired and/or wireless communication.

The network unit 300 may include a wired/wireless Internet module for network access. Wireless Internet technologies include wireless LAN (WLAN) (Wi-Fi), wireless broadband (Wibro), world interoperability for microwave access (Wimax), high speed downlink packet access (HSDPA), and the like. As a wired Internet technology, XDSL (Digital Subscriber Line), FTTH (Fibers to the home), PLC (Power Line Communication), and the like may be used.

In addition, the network unit 300 includes a short-distance communication module and is located at a relatively short distance from the computing device to transmit/receive data with an electronic device including a short-distance communication module. As a short range communication technology, Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, or the like may be used.

2 is a flowchart of a method for predicting the occurrence of depression using an artificial intelligence model according to an embodiment of the present invention.

Referring to FIG. 2, in the method of predicting depression occurrence using an artificial intelligence model (S400), a processor of a computing device receives big data including biological data, psychological data, and sociological data from a first server connected to the device. Depression through a receiving step (S410), a first parameter extraction step (S430) of extracting a parameter related to the occurrence of depression among the received data by the processor, and a result value obtained by inputting the extracted first parameter into the learned artificial intelligence model. A depression occurrence prediction step (S450) of predicting occurrence may be included.

In one embodiment, the data receiving step ( S410 ) may be a step in which the processor 100 of the computing device receives big data including biological data, psychological data, and sociological data from a server connected to the device.

Here, big data may include bio-signal data measured through sensors, psychological test data through surveys, social network survey data through surveys, and activity data using communication networks.

More specifically, big data may include clinical research data, survey data responding to preset questions and answers, psychological tests, user biometric data measured through sensors of electronic devices, and SNS activities used by users through communication devices. , activity data such as text messages or phone calls, and the like.

In addition, biometric data may include biosignals such as real-time heart rate, amount of exercise, amount of sleep, etc. Psychological test data is data to respond to short questions about mood, mood, and mental health status periodically or non-periodically. Network survey data may include information such as characteristics of social relationships (eg, size of personal network, type of personal network, strength of personal network, etc.), amount of phone calls, amount of social activity using GPS, etc., and SNS activity data.

In one embodiment, the first parameter extraction step ( S430 ) may be a step in which the processor 100 extracts parameters related to the occurrence of depression among the received data.

More specifically, in the first parameter extraction step (S430), parameters including factors directly or indirectly related to the occurrence of depression may be primarily extracted from the big data received from the server.

Figure 3 is a flow chart of the step of predicting the occurrence of depression according to an embodiment of the present invention. Referring to FIG. 3, the depression occurrence prediction step (S450) may be a step of predicting the occurrence of depression through a result obtained by inputting the first parameter extracted in the first parameter extraction step (S430) to the learned artificial intelligence model. there is.

More specifically, the depression occurrence prediction step (S450) quantifies the first parameter according to the contribution of depression occurrence through decision tree analysis and regression analysis on the extracted first parameter, A second parameter extraction step of extracting a parameter with a high contribution among the digitized first parameters and a depression occurrence prediction step of predicting the occurrence of depression through a result value obtained by inputting the extracted second parameter to an artificial intelligence model. can

In addition, weights may be assigned to parameters having a high contribution to the occurrence of depression, and the prediction of occurrence of depression may increase as parameters having a high contribution are included.

4 is a diagram showing parameters extracted according to the contribution of depression occurrence according to an embodiment of the present invention, and FIG. 5 is a diagram showing an ensemble learning model according to an embodiment of the present invention.

Referring to FIG. 4, parameter extraction can perform dimension reduction through the influence analysis between input data and predictive shape depression occurrence in order to simplify the model and increase feasibility, and for dimension reduction, a decision tree (decision tree) Regression analysis can be performed based on tree. As a result, it is possible to quantify the contribution of all input data and parameters to whether they are related to the occurrence of depression, rank the quantified results, and select parameters according to the number of parameters set in advance by the user. can

In addition, referring to FIG. 5 , the prediction model may be learned and evaluated by applying the selected second parameter to a machine learning learning technique. As machine learning learning methods, decision tree (DT), logistic regression (LR), k-nearest neighbor (KNN), support vector machine (SVM), and ensemble learning (EL) can be used to design and learn predictive models. In addition, the artificial intelligence model of the present invention may be characterized by using an ensemble learning (EL)-based prediction model. The ensemble learning model is composed of several weak learners (week learners) as shown in FIG. 5, and the final result is a machine learning method in which the overall average of all weak learner results is determined. In this study, a weak learner was constructed with 30 DT classifiers for model optimization and high performance.

Here, the artificial intelligence model, as a prediction algorithm, may include an ensemble learning (EL) method. In addition, the artificial intelligence model may use a plurality of machine learning algorithms for the second parameter having a high contribution to the occurrence of depression, and determine the overall average value of the respective result values as the final value. The accuracy of prediction can be increased by using the above method.

In addition, the artificial intelligence model may be characterized by learning information related to the occurrence of depression using the aforementioned big data.

6 is a diagram showing a result of predicting the occurrence of depression using an artificial intelligence model according to an embodiment of the present invention.

Referring to FIG. 6, the performance of the depression prediction model of the depression occurrence prediction method using an artificial intelligence model was evaluated using precision, recall, sensitivity, F1-score and AUC values, As a result, precision of 84.% and recall of 100.0% were shown, the F1-score was 91.5, and the AUC value was 97.0, indicating high performance.

The method for predicting the occurrence of depression using an artificial intelligence model according to an embodiment of the present invention may also be implemented in the form of a computer-readable recording medium recording a program for execution on a computer. A computer-readable recording medium may be any available medium that can be accessed by a computer, and may include both volatile and nonvolatile media, removable and non-removable media. Also, computer readable recording media may include computer storage media. Computer storage media may include both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

Exemplary modules, steps, or combinations thereof related to the embodiments described in this specification may be implemented by electronic hardware (digital design designed by coding, etc.), software (various types of applications including program instructions), or a combination thereof. can be implemented Which form of hardware and/or software is implemented may vary depending on design constraints imposed on the user terminal.

One or more of the configurations described herein may be stored in memory as computer program instructions, which may execute the methods described herein centered on a digital signal processor. Connection examples between components specified with reference to the drawings attached to this specification are merely illustrative, and at least some of them may be omitted, and conversely, additional components as well as these components may be further included.

The above description is only illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention.

Therefore, the embodiments disclosed in this specification are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the present invention is not limited by these embodiments.

The protection scope of the present invention should be interpreted according to the claims, and all technical ideas within the equivalent range should be understood to be included in the scope of the present invention.

Claims (6)

In a method for predicting the occurrence of depression using an artificial intelligence model by a computing device,
a data receiving step in which a processor of the computing device receives big data including biological data, psychological data, and sociological data from a server connected to the device;
a first parameter extraction step of extracting, by the processor, a parameter related to occurrence of depression among the received data; and
Including a depression occurrence prediction step of predicting the occurrence of depression through a result value obtained by inputting the extracted first parameter into the learned artificial intelligence model,
Depression prediction method using artificial intelligence model.
According to claim 1,
The big data,
Characterized in that it includes bio-signal data measured through sensors, psychological test data through surveys, social network survey data through surveys, and activity data using communication networks.
Depression prediction method using artificial intelligence model.
According to claim 1,
The depression occurrence prediction step,
Through decision tree analysis and regression analysis on the extracted first parameter, the first parameter is quantified according to the occurrence of depression, and the parameter with a high contribution among the digitized first parameters A second parameter extraction step of extracting ; and
Including a depression occurrence prediction step of predicting the occurrence of depression through a result value obtained by inputting the extracted second parameter into an artificial intelligence model,
Depression prediction method using artificial intelligence model.
According to claim 3,
The artificial intelligence model,
As a prediction algorithm, it is characterized by using an ensemble learning (EL) method,
Depression prediction method using artificial intelligence model.
According to claim 4,
The artificial intelligence model,
Characterized in that the second parameter is determined by using a plurality of machine learning algorithms as the final value, the average value of the entire result values, respectively,
Depression prediction method using artificial intelligence model.
A computer-readable recording medium in which a program for executing the method according to any one of claims 1 to 5 in a computer is recorded.

Images