KR102539760B1 - Apparatus for predicting clinical evaluation score using a game for ADHD diagnosis and method thereof - Google Patents

Abstract

The present invention relates to an apparatus and method for predicting clinical evaluation scores using a game for diagnosing ADHD.
The apparatus for predicting clinical evaluation scores according to the present invention includes a game providing unit providing a game developed for training of working memory and attention, a personal information input unit receiving personal information of a measurement subject to receive clinical evaluation for ADHD, and a measurement subject's personal information input unit. A game information acquisition unit that acquires game result information according to playing a selected game, a variable extractor that extracts a cognitive predictor variable from the obtained game result information, and inputs the personal information and cognitive predictor variables to a predicted model that has been trained. and a prediction unit for predicting the Wechsler intelligence scale for children (WISC-V) score.
As described above, according to the present invention, children can perform content on their own without restrictions according to time and place, and clinical evaluation scores can be predicted without being evaluated by experts with professional qualifications such as clinical psychologists, so that ADHD can be detected early. have an effect

Description

Apparatus for predicting clinical evaluation score using a game for ADHD diagnosis and method thereof}

The present invention relates to an apparatus and method for predicting clinical evaluation scores using a game for diagnosing ADHD, and more specifically, processing speed scores according to reaction time, number of touches, etc. of children playing games using an AI algorithm, and tasks It relates to a clinical evaluation score prediction device and method for predicting memory scores and perceptual reasoning scores.

Attention-Deficit/Hyperactivity Disorder (ADHD) is characterized by a lack of attention, distraction, hyperactivity, and impulsivity.

Symptoms vary from person to person, and when hyperactivity/impulsiveness and attention deficit appear together (ADHD-C), when only attention deficit symptoms appear (ADHD-PI), when only hyperactivity/impulsive symptoms appear mainly (ADHD-PH) ) is divided into Symptoms of ADHD　The diagnostic criteria for ADHD are not simple. Diagnosis is made through a comprehensive judgment by a specialist on current symptoms, past history, presence of comorbidities, and whether diseases other than ADHD caused attention deficit.

The IQ of patients with attention deficit hyperactivity (hyperactivity) disorder (ADHD) is measured normally. Therefore, most people do not recognize attention deficit hyperactivity (hyperactivity) disorder (ADHD) as one of the diseases and regard poor social skills as a problem of individual personality or will.

In addition, even if attention deficit hyperactivity (hyperactivity) disorder (ADHD) is recognized, it is burdensome to visit the psychiatric department, and furthermore, the diagnosis is not made properly. There is a problem that the number of patients that can be accommodated is small due to the tests that lead to, and the disease classification may vary depending on the clinician interpreting the test results.

The background technology of the present invention is disclosed in Republic of Korea Patent Registration No. 10-2438580 (2022.09.01. Notice).

As described above, according to the present invention, to provide a clinical evaluation score prediction device and method for predicting processing speed scores, working memory scores, and perceptual reasoning scores according to the reaction time, number of touches, etc. of children playing games using AI algorithms It is for

According to an embodiment of the present invention to achieve this technical problem, a clinical evaluation score prediction device for diagnosing ADHD is provided by a game provider providing a game developed for training working memory and attention, and a measurement subject who wants to receive a clinical evaluation for ADHD. A personal information input unit for receiving personal information of, a game information acquisition unit for acquiring game result information according to the game selected by the subject of measurement, a variable extractor for extracting a cognitive predictor variable from the obtained game result information, and the individual It includes a prediction unit that predicts the Wechsler intelligence scale for children-V (WISC-V) score by inputting information and cognitive predictors into the predicted model after learning has been completed.

The method may further include a result providing unit that determines whether an additional test for diagnosing ADHD is required using the predicted score and provides a result of the determination.

A data collection unit that collects Wechsler Child Intelligence Test (WISC-V) scores and game result information of multiple patients diagnosed with ADHD, and the Wechsler Child Intelligence Test (WISC-V) using the collected game result information as input data It may further include a learning unit for learning a pre-learned prediction model using scores as output data.

The game result information includes the dominant hand selection, content name, practice game or evaluation game classification, content play time, total number of questions performed, correct answers, game difficulty, user touch time, user touch count, content speed within the question, problem It may include at least one of a correct answer, a location of a correct answer to a question, a location of a user's correct answer, an item selected by a user, the number of pauses, whether the game is interesting, and a perceived level of difficulty of the game.

The variable extractor processes the game result information to determine the correct answer rate, the average reaction time, the standard deviation of the reaction time, the maximum reaction time, the minimum reaction time, the average/standard deviation/maximum/minimum value of the response time when correct, and the wrong answer day. average/standard deviation/maximum/minimum value, first touch average/standard deviation/maximum/minimum value, first touch average/standard deviation/maximum/minimum value for correct answer, first touch for incorrect answer Time average / standard deviation / maximum value / minimum value, total number of touches, total number of left touches, total number of right touches, average difficulty / standard deviation / maximum value, average speed / standard deviation / maximum value, effective touch value, when correct At least one cognitive predictor variable may be extracted from among problem time average/standard deviation/maximum value/minimum value, problem time average/standard deviation/maximum value/minimum value when an answer is incorrect, target stimulus classification value, and careless impulse value.

The learning unit performs a regression analysis between the game result information and the Wechsler child intelligence test (WISC-V) score, and among the plurality of cognitive predictor variables, the standard deviation of the alignment time, the number of user touches in the case of an incorrect answer, and the maximum utility of the target stimulus in the case of a correct answer The prediction model may be trained to predict the processing speed score by assigning weights to the value, the maximum response time value in case of a correct answer, and the maximum value of the first touch in case of a correct answer, respectively.

The learning unit performs a regression analysis between the game result information and the Wechsler child intelligence test (WISC-V) score, and among the plurality of cognitive predictors, average reaction time, average reaction time when answering incorrectly, maximum value of user final touch time, user The prediction model may be trained to predict the spatiotemporal score by assigning weights to the total number of touches and the number of impulsive touches.

The learning unit performs a regression analysis between the game result information and the Wechsler child intelligence test (WISC-V) score, and among the plurality of cognitive predictors, the maximum value of alignment time when correct, the standard deviation of alignment time when correct, and the standard deviation of alignment time when correct The predictive model may be trained to predict the working memory score by assigning weights to the average value of target stimulus efficacy, the standard deviation of target stimulus utility when correct answer is given, and the maximum reaction time value when correct answer is given.

The prediction unit may obtain processing speed scores, spatiotemporal scores, and working memory scores from the prediction model by inputting personal information and cognitive predictive variables of a measurement subject to be clinically evaluated into a prediction model.

In addition, according to an embodiment of the present invention, the clinical evaluation score prediction method using the clinical evaluation score prediction device includes the steps of providing a game developed for working memory and attention training, and personal information of a measurement subject who wants to receive a clinical evaluation for ADHD. Receiving input, obtaining game result information according to the game selected by the measurement subject, extracting cognitive predictors from the obtained game result information, and predicting that the personal information and cognitive predictors have been learned It includes a step of predicting the Wechsler intelligence scale for children-V (WISC-V) score by inputting it to the model.

As described above, according to the present invention, children can perform content on their own without restrictions according to time and place, and clinical evaluation scores can be predicted without being evaluated by experts with professional qualifications such as clinical psychologists, so that ADHD can be detected early. have an effect

1 is a configuration diagram illustrating an apparatus for predicting clinical evaluation scores according to an embodiment of the present invention.
2 is an exemplary diagram for explaining an animal matching game developed for training a subject's working memory.
3 is an exemplary diagram for explaining a glass bridge game for attention training of a subject.
4 is an exemplary diagram for explaining a glass bridge game to find a friend for attention training of a subject.
5 is an exemplary diagram for explaining a vortex spacecraft for attention training of a subject.
6 is an exemplary diagram for explaining a space travel game for attention training of a subject.
7 is a flowchart illustrating a clinical evaluation score prediction method using the clinical evaluation score prediction apparatus according to an embodiment of the present invention.
FIG. 8 is an exemplary view for explaining step S730 shown in FIG. 7 .

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. In this process, the thickness of lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of description.

In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of a user or operator. Therefore, definitions of these terms will have to be made based on the content throughout this specification.

Hereinafter, an apparatus for predicting clinical evaluation scores using a game for diagnosing ADHD according to an embodiment of the present invention will be described in more detail with reference to FIG. 1 .

1 is a configuration diagram illustrating an apparatus for predicting clinical evaluation scores according to an embodiment of the present invention.

As shown in FIG. 1, the clinical evaluation score prediction apparatus 100 according to an embodiment of the present invention includes a game providing unit 110, a data collection unit 120, a learning unit 130, and a personal information input unit 140. , It includes a game information acquisition unit 150, a variable extraction unit 160, a prediction unit 170 and a result providing unit 180.

First, the game providing unit 110 provides various contents developed for working memory and attention training in the form of a game.

The provided games can be divided into practice games and evaluation games, depending on the level of difficulty.

The data collection unit 120 collects Wechsler Child Intelligence Test (WISCW) scores and game result information of a plurality of patients diagnosed with ADHD.

In general, the overall scale of the Wechsler Child Intelligence Scale (WISC-V) consists of verbal comprehension, visuospatial, fluid reasoning, working memory, and processing speed. In the embodiment of the present invention, since the scores for processing speed, space-time, and working memory are predicted among the five domains, it is okay for the data collection unit 120 to collect only the scores for processing speed, space-time, and working memory.

The learning unit 130 builds a predictive model for predicting the clinical evaluation score, and trains the constructed predictive model using the ADHD patients' Wechsler Child Intelligence Test (WISC-V) scores and game result information.

The personal information input unit 140 receives personal information of a subject for whom ADHD clinical evaluation is to be performed. Here, the personal information includes at least one of age, gender, smartphone ownership, smartphone usage time, and smartphone game presence.

The game information acquisition unit 150 acquires game result information as the measurement subject performs the selected game.

The variable extractor 160 extracts a plurality of cognitive predictor variables related to cognitive characteristics by processing the obtained game result information.

To elaborate, the variable extractor 160 extracts a plurality of cognitive predictor variables related to processing speed, extracts a plurality of cognitive predictor variables related to working memory, and extracts a plurality of cognitive predictor variables related to perceptual reasoning.

The prediction unit 170 inputs the personal information and cognitive predictive variables of the subject to be measured into the predicted model to obtain predicted processing speed scores, spatiotemporal scores, and working memory scores from the predicted model.

The result provider 180 sets reference values for the processing speed score, spatio-temporal score, and working memory score, and compares the score obtained through the prediction unit 170 with the preset reference value to determine whether additional tests for diagnosing ADHD are needed. judge Then, the result providing unit 180 transmits the determined result to the terminal of the measurement target.

Hereinafter, a clinical evaluation score prediction method using the clinical evaluation score prediction apparatus 100 according to an embodiment of the present invention will be described in detail with reference to FIGS. 2 to 8 .

First, the clinical evaluation score prediction apparatus 100 according to an embodiment of the present invention provides a game developed for working memory and attention training through an app or a web page.

2 is an exemplary diagram for explaining an animal matching game developed for training a subject's working memory.

As shown in Figure 2, multiple pairs of animals are hidden through the log box, and then the hidden logs are touched to match hidden identical animals. At this time, the number of animals to be memorized varies according to the difficulty of the game.

3 is an exemplary diagram for explaining a glass bridge game for attention training of a subject.

As shown in FIG. 3, three legs are created, and the location of the animal is changed by touching the arrows formed on both sides of the legs. Comparing the approaching animal and the animal to be moved, if they have the same shape, move the corresponding animal to the right or left to match the position of the approaching animal. At this time, the speed of the approaching animal is changed according to the difficulty of the game.

4 is an exemplary diagram for explaining a glass bridge game to find a friend for attention training of a subject.

As shown in FIG. 4 , three legs are created, three animals appearing close to the user are compared with three approaching animals, and animals of the same type (shape) are matched. In the above game, touch the left arrow to change the position of the number 1 animal and the number 2 animal, and touch the right arrow to change the position of the number 2 animal and the number 3 animal to align the animals. At this time, the speed of the approaching animal is changed according to the difficulty of the game.

5 is an exemplary diagram for explaining a vortex spacecraft for attention training of a subject.

As shown in Figure 5, two spaceships appear diagonally on the six rails formed inside the cylinder. Then, when problem boxes appear on both sides of the screen, change the location of the two spaceships to the location where the problem boxes are.

At this time, by touching the button located at the bottom of the screen, a pair of spacecraft located diagonally moves counterclockwise or clockwise, and the speed at which the question box appears and disappears varies depending on the difficulty of the game.

6 is an exemplary diagram for explaining a space travel game for attention training of a subject.

As shown in FIG. 6, rails are formed on the left and right sides of each spacecraft, and when a problem box appears during flight, the spacecraft is relocated to the rail where the corresponding problem box is located.

The spacecraft is moved to the left or right by touching the button located at the bottom, and each spacecraft proceeds with the game by solving problems with time differences. At this time, the speed at which the problem box appears and disappears varies according to the difficulty of the game.

As described above, in the embodiment of the present invention, various games are provided to stimulate children's interest and train working memory and attention.

7 is a flowchart illustrating a clinical evaluation score prediction method using the clinical evaluation score prediction apparatus according to an embodiment of the present invention.

As shown in FIG. 7 , the clinical evaluation score is predicted using the provided game according to an embodiment of the present invention, and the method for predicting the clinical evaluation score includes the step of learning the predictive model and the clinical evaluation using the learned predictive model. It consists of predicting the score.

First, looking at the step of learning the predictive model, the data collection unit 120 collects Wechsler Child Intelligence Test (WISC-V) scores and game result information of a plurality of patients diagnosed with ADHD (S710).

The data collection unit 120 collects Wechsler Child Intelligence Scale (WISC-V) scores of a plurality of patients diagnosed with ADHD. At this time, the collected Wechsler Child Intelligence Scale (WISC-V) score includes at least one of a processing speed score, a spatiotemporal score, and a working memory score.

Further, the data collecting unit 120 collects game result information obtained as a plurality of patients diagnosed with ADHD play the games shown in 3 to 6 above.

Here, the game result information includes the dominant hand selection, content name, practice game or evaluation game classification, content play time, total number of questions performed, correct answers, game difficulty, user touch time, user touch count, content speed within the question, problem It includes at least one of a correct answer, a location of a correct answer to a problem, a location of a user's correct answer, an item selected by a user, the number of pauses, whether the game is interesting, and a perceived level of difficulty of the game.

Next, the learning unit 130 analyzes the collected Wechsler Child Intelligence Test (WISC-V) scores and game result information to extract cognitive characteristic variables to be weighted (S720).

To explain this again, the variable extractor 160 processes the collected game result information to extract a cognitive predictor variable.

Here, the cognitive predictors are percent correct, average response time, standard deviation of reaction time, maximum response time, minimum response time, mean/standard deviation/maximum/minimum value of response time for correct answers, mean/standard deviation/reaction time for incorrect answers. Max/Min value, first touch time average/standard deviation/max/min value, first touch time average/standard deviation/max/min value for correct answer, first touch time average/standard deviation/max value for incorrect answer / minimum value, total number of touches, total number of left touches, total number of right touches, average difficulty / standard deviation / maximum value, average speed / standard deviation / maximum value, valid touch value, average problem time when correct / standard deviation / maximum It includes at least one of value/minimum value, problem time average/standard deviation/maximum value/minimum value in case of incorrect answer, target stimulus classification value, and inattention impulse value.

Next, the learning unit 130 performs a regression analysis on the correlation between the Wechsler Child Intelligence Test (WISC-V) score and cognitive predictors using machine learning, which is of high importance in predicting processing speed scores, spatiotemporal scores, and spatiotemporal scores. A plurality of cognitive characteristic variables are extracted.

To explain this again, the learning unit 130 uses the standard deviation of sorting time as important variables in predicting the processing speed score, the number of user touches in case of an incorrect answer, the maximum value of target stimulus utility in case of a correct answer, the maximum value of reaction time in case of a correct answer, and the correct answer When , the maximum value of the first touch is extracted.

In addition, the learning unit 130 extracts the average reaction time, the average reaction time for incorrect answers, the maximum value of the user's final touch time, the total number of user touches, and the number of impulsive touches as important variables in predicting the spatiotemporal score.

In addition, the learning unit 130 determines the maximum value of sorting time in case of correct answer, the standard deviation of sorting time in case of correct answer, the average value of target stimulus utility in case of correct answer, the standard deviation of target stimulus utility in case of correct answer, and When the answer is correct, the maximum reaction time is extracted.

When step S720 is completed, the learning unit 130 trains the predictive model using the Wechsler Child Intelligence Test (WISC-V) score and cognitive predictors (S730).

FIG. 8 is an exemplary view for explaining step S730 shown in FIG. 7 .

As shown in FIG. 8, the learning unit 130 uses the game result information, a plurality of cognitive characteristic variables extracted from among a plurality of cognitive predictors, and the subject's Wechsler Child Intelligence Test (WISC-V) score to form a predictive model. to predict the Wechsler Child Intelligence Scale (WISC-V) score.

Meanwhile, in an embodiment of the present invention, weights are assigned to a plurality of cognitive characteristic variables extracted as important variables in predicting the processing speed score, the spatiotemporal score, and the working memory score in order to increase the accuracy according to the prediction of the clinical evaluation score.

After completing the learning of the predictive model in steps S710 to 730, the clinical evaluation score prediction apparatus 100 acquires personal information and game result information of a subject to be evaluated for ADHD clinically (S740).

The measurement subject inputs personal information while accessing an app or web page provided by the clinical evaluation score prediction device 100 through a terminal.

Here, the personal information includes at least one of age, gender, smartphone ownership, smartphone usage time, and smartphone game status of the measurement subject.

Next, the subject of measurement selects one of a plurality of game contents provided by the clinical evaluation score prediction device 100 and plays a practice game or an evaluation game.

Then, the game information obtaining unit 150 acquires the measured game result information while the measurement subject is playing the game.

When step S740 is completed, the variable extraction unit 160 extracts a cognitive predictor variable from the game result information obtained from the game information acquisition unit 150 (S750).

Here, the cognitive predictors are percent correct, average response time, standard deviation of reaction time, maximum response time, minimum response time, mean/standard deviation/maximum/minimum value of response time for correct answers, mean/standard deviation/reaction time for incorrect answers. Max/Min value, first touch time average/standard deviation/max/min value, first touch time average/standard deviation/max/min value for correct answer, first touch time average/standard deviation/max value for incorrect answer / minimum value, total number of touches, total number of left touches, total number of right touches, average difficulty / standard deviation / maximum value, average speed / standard deviation / maximum value, valid touch value, average problem time when correct / standard deviation / maximum It includes at least one of value/minimum value, problem time average/standard deviation/maximum value/minimum value in case of incorrect answer, target stimulus classification value, and inattention impulse value.

The personal information obtained through the personal information input unit 140 and the cognitive predictor variables obtained through the variable extractor 160 are transmitted to the predictor 170 .

Then, the prediction unit 170 inputs the received personal information and cognitive predictor variables into the predicted model for which learning has been completed to obtain Wechsler child intelligence test (WISC-V) scores (S760).

The prediction unit 170 inputs the received personal information and cognitive predictor variables into the predicted model after learning has been completed. Then, the predictive model outputs processing speed scores, spatiotemporal scores, and working memory scores in response to the input personal information and cognitive predictors.

Finally, the result providing unit 180 determines whether an additional test for diagnosing ADHD is necessary and provides the determined result (S770).

In general, a Wechsler Child Intelligence Scale (WISCW) score between 85 and 115 is considered normal. However, in the present invention, the Wechsler Child Intelligence Test (WISC-V) score is not inferred through the test results in five areas, but only the scores for processing speed, spatio-temporal and working memory are predicted, so the result providing unit 180 is ADHD A reference range is established to determine the need for additional tests for diagnosis.

Next, the result providing unit 180 determines whether the score predicted through the predictive model corresponds to the reference range. And, if the predicted score is smaller than the lower limit of the reference range, the result providing unit 180 transmits a signal to the terminal of the corresponding measurement subject to perform an additional test for diagnosing ADHD.

As described above, the clinical evaluation score prediction device according to the present invention can perform contents on its own without time and place limitations, and can predict clinical evaluation scores without being evaluated by a professionally qualified expert such as a clinical psychologist, thereby helping to prevent ADHD. It has the effect of early detection.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is only exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. will be. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the claims below.

100: clinical evaluation score prediction device
110: game provision unit
120: data collection unit
130: learning unit
140: personal information input unit
150: game information acquisition unit
160: variable extraction unit
170: prediction unit
180: result providing unit

Claims (18)

In the clinical evaluation score predictive device for diagnosing ADHD,
A game providing unit that provides games developed for training working memory and attention;
A personal information input unit that receives personal information of a subject to be measured for clinical evaluation of ADHD;
A game information acquisition unit that acquires game result information as the measurement subject performs a game selected by the subject;
A variable extractor for extracting a cognitive predictor variable from the obtained game result information;
A prediction unit that predicts the Wechsler intelligence scale for children-V (WISC-V) score by inputting the personal information and cognitive predictor variables into the predicted model after learning has been completed;
A data collection unit that collects Wechsler Child Intelligence Test (WISC-V) score and game result information of multiple patients diagnosed with ADHD, and
A learning unit for learning a pre-learned prediction model using the collected game result information as input data and Wechsler Child Intelligence Test (WISCW) scores as output data;
The learning unit,
A clinical evaluation score prediction device that trains the prediction model to predict a processing speed score by performing a regression analysis between the game result information and the Wechsler child intelligence test (WISC-V) score and assigning weights to the plurality of cognitive predictors, respectively. . According to claim 1,
The clinical evaluation score prediction apparatus further comprising a result providing unit determining whether an additional test for diagnosing ADHD is required using the predicted score and providing a result of the determination. delete According to claim 1,
The game result information,
Selection of dominant hand, name of content, classification of practice game or evaluation game, time of content play, total number of questions performed, presence or absence of correct answer, game difficulty, user touch time, number of user touches, speed of content within the question, correct answer to question, position of correct answer to question, A device for predicting clinical evaluation scores that includes at least one of a user's correct answer position, an item selected by the user, the number of pauses, interest in the game, and a perceived level of difficulty of the game. According to claim 1,
The variable extraction unit,
By processing the game result information,
Percentage of correct answers, average response time, standard deviation of response time, maximum value of response time, minimum value of reaction time, mean/standard deviation/maximum/minimum value of response time for correct answers, average/standard deviation/maximum/minimum value of response times for incorrect answers, First touch time mean/standard deviation/maximum/min value, correct first touch time mean/standard deviation/maximum/min value, first touch time mean/standard deviation/maximum/min value for incorrect answers, total touches Number of times, total number of left touches, total number of right touches, average difficulty / standard deviation / maximum value, speed average / standard deviation / maximum value, effective touch value, average / standard deviation / maximum value / minimum value of problem time when correct, wrong answer A clinical evaluation score prediction device that extracts at least one cognitive predictor variable among problem time mean/standard deviation/maximum/minimum value, target stimulus classification value, and inattention impulse value when . According to claim 1,
The learning unit,
Regression analysis is performed between the game result information and the Wechsler child intelligence test (WISC-V) score, and among the plurality of cognitive predictors, the standard deviation of the alignment time, the number of user touches in case of an incorrect answer, the maximum value of target stimulus efficacy in case of correct answer, and the date of correct answer The clinical evaluation score prediction apparatus for training the prediction model to predict the processing speed score by assigning weights to the maximum response time value and the first touch maximum value when the answer is correct. According to claim 1,
The learning unit,
Regression analysis was performed between the game result information and the Wechsler Child Intelligence Test (WISC-V) score, and among the plurality of cognitive predictors, the average reaction time, the average reaction time when answering incorrectly, the maximum value of the user's final touch time, the total number of user touches, and An apparatus for predicting clinical evaluation scores that assigns weights to the number of impulsive touches and trains the predictive model to predict spatiotemporal scores. According to claim 1,
The learning unit,
Regression analysis is performed between the game result information and the Wechsler Child Intelligence Test (WISC-V) score, and among the plurality of cognitive predictors, the maximum value of sorting time when correct, the standard deviation of sorting time when correct, and the average value of target stimulus efficacy when correct , A clinical evaluation score prediction device that trains the predictive model to predict a working memory score by assigning weights to the standard deviation of the target stimulus utility when the answer is correct and the maximum value of the reaction time when the answer is correct. According to claim 1,
The prediction unit,
A clinical evaluation score prediction device that obtains processing speed scores, spatiotemporal scores, and working memory scores from the prediction model by inputting personal information and cognitive predictors of a measurement subject to be subjected to clinical evaluation into a prediction model. In the clinical evaluation score prediction method for diagnosing ADHD using a clinical evaluation score prediction device,
Collecting Wechsler Child Intelligence Test (WISC-V) score and game result information of a plurality of patients diagnosed with ADHD;
Training a pre-learned predictive model using the collected game result information as input data and Wechsler Child Intelligence Test (WISC-V) scores as output data;
providing a game developed for training working memory and attention;
Step of receiving the personal information of the subject to be measured for clinical evaluation of ADHD;
Obtaining game result information according to the measurement subject performing the selected game;
extracting a cognitive predictor variable from the obtained game result information; and
Predicting a Wechsler intelligence scale for children-V (WISC-V) score by inputting the personal information and cognitive predictor variables into a predicted model after learning has been completed,
The step of learning the predictive model,
Clinical evaluation score prediction method for learning to predict a processing speed score by assigning a weight to each of the plurality of cognitive predictors by performing a regression analysis between the game result information and the Wechsler child intelligence test (WISC-V) score . According to claim 10,
The clinical evaluation score prediction method further comprising determining whether an additional test for diagnosing ADHD is required using the predicted score and providing a result of the determination. delete According to claim 10,
The game result information,
Selection of dominant hand, name of content, classification of practice game or evaluation game, time of content play, total number of questions performed, presence or absence of correct answer, game difficulty, user touch time, number of user touches, speed of content within the question, correct answer to question, position of correct answer to question, A clinical evaluation score prediction method that includes at least one of the user's correct answer location, the user's selected item, the number of pauses, whether the game is interesting, and the degree of difficulty experienced by the game. According to claim 10,
Extracting the cognitive predictor variable,
By processing the game result information,
Percentage of correct responses, average response time, standard deviation of response time, maximum response time, minimum response time, average/standard deviation/maximum/minimum value of response time for correct answers, average/standard deviation/maximum/minimum response times for incorrect answers, First touch time mean/standard deviation/maximum/min value, correct first touch time mean/standard deviation/maximum/min value, first touch time mean/standard deviation/maximum/min value for incorrect answers, total touches Number of times, total number of left touches, total number of right touches, difficulty average / standard deviation / maximum value, speed average / standard deviation / maximum value, effective touch value, problem time average / standard deviation / maximum value / minimum value when correct, wrong answer A clinical evaluation score prediction method that extracts at least one cognitive predictor among problem time mean/standard deviation/maximum/minimum value, target stimulus classification value, and inattention impulse value when . According to claim 10,
The step of learning the predictive model,
Regression analysis is performed between the game result information and the Wechsler child intelligence test (WISC-V) score, and among the plurality of cognitive predictors, the standard deviation of the alignment time, the number of user touches in case of an incorrect answer, the maximum value of target stimulus efficacy in case of correct answer, and the date of correct answer A clinical evaluation score prediction method in which the prediction model learns to predict the processing speed score by assigning weights to the maximum response time value and the first touch maximum value when the answer is correct. According to claim 10,
The step of learning the predictive model,
Regression analysis was performed between the game result information and the Wechsler Child Intelligence Test (WISC-V) score, and among the plurality of cognitive predictors, the average reaction time, the average reaction time when answering incorrectly, the maximum value of the user's final touch time, the total number of user touches, and A method for predicting clinical evaluation scores in which the predictive model learns to predict spatiotemporal scores by assigning weights to the number of impulsive touches. According to claim 10,
The step of learning the predictive model,
Regression analysis is performed between the game result information and the Wechsler child intelligence test (WISC-V) score, and among the plurality of cognitive predictors, the maximum value of alignment time when correct, the standard deviation of alignment time when correct, and the average value of target stimulus efficacy when correct , Clinical evaluation score prediction method in which the predictive model learns to predict the working memory score by assigning weights to the standard deviation of the target stimulus utility when the answer is correct and the maximum value of the reaction time when the answer is correct. According to claim 10,
The step of predicting the Wechsler Child Intelligence Test (WISC-V) score,
A clinical evaluation score prediction method for acquiring processing speed scores, spatiotemporal scores, and working memory scores from the prediction model by inputting personal information and cognitive predictors of a measurement subject to be subjected to clinical evaluation into a prediction model.

Images