CN117769747A

CN117769747A - Method and system for diagnosing dementia by story-based question-answer assessment and story-based speech extraction

Info

Publication number: CN117769747A
Application number: CN202280053587.5A
Authority: CN
Inventors: 李准荣; 卢俞宪
Original assignee: Yimoke Co ltd
Current assignee: Yimoke Co ltd
Priority date: 2021-08-09
Filing date: 2022-08-09
Publication date: 2024-03-26
Also published as: KR102438778B1

Abstract

The present invention relates to a system for diagnosing dementia using story-based question-answer type assessment, which can rapidly and easily diagnose dementia without a professional, and rapidly and accurately diagnose dementia and screen light patients, and can easily check even elderly people having difficulty in recognizing characters by performing question-answer type assessment through voice. The system for diagnosing dementia comprises: a data input unit capable of inputting an actual story; a data memory storing text data inputted through the data input unit; a voice output device for outputting voice data for diagnosis to the outside through voice; a voice receiver for receiving an answer voice of the subject; and a controller connected to and controlling the various devices including the data input unit, and forming an adapted story, a question, and a correct answer based on the actual story, and receiving an answer of the subject, and comparing the answer with the correct answer to determine dementia.

Description

Method and system for diagnosing dementia by story-based question-answer assessment and story-based speech extraction

Technical Field

The present invention relates to a system and method for diagnosing dementia, and more particularly, to a system and method for diagnosing dementia using story-based question-answer type evaluation and story-based speech extraction, which can rapidly and easily diagnose dementia without a professional, and rapidly and accurately diagnose dementia and screen light patients, and can easily check even elderly people having difficulty in recognizing characters by question-answer type evaluation through speech or speech prompt of a story.

Background

Dementia generally refers to a series of symptoms caused by brain diseases. As dementia progresses, it affects mental capacity, actions and daily life performance, and is characterized by a lack of daily mobility due to a decline in cognitive ability.

In general, when two or more cognitive functions are significantly impaired, a doctor may diagnose dementia. Such cognitive functions include memory, language skills, information understanding, spatial skills, diagnostic power and attention. Dementia patients may experience difficulties in solving problems and controlling emotion and may experience character changes.

The exact symptoms experienced by dementia patients depend on the site of damage to the brain by the disease causing the dementia. Dementia, in which a part of the nerve cells of the brain are stopped in function, lose communication with other cells and die, is generally steadily progressed. That is, dementia gradually spreads to the brain, and the symptoms of the patient worsen with the passage of time.

Dementia is a typical neurodegenerative brain disease associated with age, with about 5-10% of the prevalence among elderly people over 65 years of age worldwide, and most dementia patients exhibit symptoms such as progressive cognitive dysfunction, hallucinations, delusions, and disability.

In the case of Alzheimer's disease, which is most typical of dementia, in nerve cells of the cerebral cortex, a waste mass called amyloid beta, which is a waste material presumably causes necrosis of nerve cells, is observed in the nerve fiber bundles and around the brain cells.

This Alzheimer's disease accounts for about 70% of all dementias, the most common disease, and recently it is estimated that about 60 thousands of Alzheimer's disease patients will occur annually in China.

Alzheimer's disease is a representative degenerative dementia disease, which causes decline of cognitive function and daily life skills through degeneration of brain nerve cells, is a dire disease which even causes death, and brings great pain to patients and great trouble to the families of care patients.

The diagnosis of dementia is to confirm the impairment of daily life and social activities caused by cognitive dysfunction through detailed medical history collection and evaluation, and to check the presence or absence of cerebrovascular diseases, brain atrophy, etc. through brain functional imaging, thereby confirming dementia.

In the early stage of dementia, it is difficult to distinguish from senile amnesia, so that neuropsychological examination is performed to comprehensively evaluate memory and language ability, computing ability, spatiotemporal perceptibility, diagnostic ability, etc. to diagnose dementia. In order to diagnose such dementia, further accurate examination is required based on interviews with patients/guardians and various information acquired through screening examination. Further precision examinations include neuropsychological examinations (SNSB), blood examinations, or various types of brain imaging examinations (CT, MRI, PET), and the like.

As an example of further precision examination, magnetic resonance imaging (magnetic resonance imaging, MRI) imaging may play an important role in distinguishing dementia types. It is possible to diagnose whether or not dementia close to Alzheimer's disease or dementia close to vascular dementia by this examination, and it is also possible to provide some help in judging whether or not dementia is caused by other diseases.

However, the prior art as described above has the following problems.

Dementia such as Alzheimer's disease is an irreversible disease, and a treatment method capable of completely curing is not yet developed, so that the dementia is diagnosed as early as possible and the progress of the dementia is delayed as an optimal scheme, but the following problems still exist: dementia diagnosis by means of detailed medical history collection and evaluation of a professional doctor is a factor that prevents rapid diagnosis of many patients due to deficiency of the professional doctor, and increases the cost of screening dementia patients in the national range.

In addition, the brain imaging diagnosis conventionally used for dementia diagnosis has the following problems: although the equipment is expensive and delicate, it can be confirmed only in cases where brain diseases or brain atrophy have progressed considerably, and thus it is difficult to use for the purpose of early diagnosis of dementia.

In addition, dementia patients are often elderly, illiterate or low in education level and cannot recognize characters well, so there are problems in that the examination itself using a questionnaire is difficult, and a lot of time and labor are required for the examination.

Therefore, there is an urgent need to develop an appropriate and breakthrough technique capable of simply and rapidly diagnosing dementia early.

Disclosure of Invention

Technical problem to be solved by the invention

Accordingly, the present invention has been made to solve the above-mentioned problems occurring in the prior art.

The present invention provides a system and method for diagnosing dementia using story-based question-answer assessment, which can rapidly and easily diagnose dementia without a professional, rapidly and accurately diagnose dementia early and screen light patients, and easily check even elderly people with difficulty in recognizing characters by performing question-answer assessment through voice.

Technical proposal

To achieve the above object, a system for diagnosing dementia using story-based questionnaire assessment according to the present invention comprises: a data input device capable of inputting text data of an actual story; a data memory which stores character data inputted through the data input unit and which is capable of storing dementia information of each subject; a voice output unit configured to output voice data for diagnosis to the outside so that the subject can hear the data; a voice receiver for receiving a response uttered by the subject; and a controller connected to the data input unit, the data storage unit, the voice output unit, and the voice receiver, the controller including a story data forming section for automatically forming adapted story data, question data, and correct answer data based on an actual story input in the form of text data, and including an answer data forming section for receiving an answer of a subject and forming answer data, and including a dementia judging section for confirming a wrong answer rate by comparing and analyzing the correct answer data and the answer data, and judging whether dementia is present and/or judging a dementia state according to the degree of the wrong answer rate, the controller including a control section for controlling the data storage unit, the voice output unit, and the voice receiver so as to sequentially output adapted story data and question data, and receiving an answer of the subject to the output question, and further storing dementia information of each of the judged subjects; the actual story is composed of a plurality of sentences as a story form centering on a person whose subject is known and an event related to the person, and the adapted story data is formed by: extracting a part of words in noun form included in the actual story and changing the extracted words in noun form to words in other noun form of the same or similar type to form an adapted story, and then converting the adapted story into voice data, the question data being data formed by: taking words in the form of altered nouns included in the adaptation story as correct answers, and forming a plurality of questions related to the adaptation story, and converting the questions into data of voice data, the correct answer data being data formed by: extracting a correct answer corresponding to the question, and converting the extracted correct answer into voice data, the answer data being voice data formed by: answers to the subject of the question received and stored by the speech receiver.

In addition, the "system for diagnosing dementia using story-based speech extraction" according to the present invention includes: a data memory storing voice data for diagnosis including story data for diagnosing dementia, and storing patient voice feature data obtained by comparing and analyzing voice features of a normal person and voice features of a dementia patient to extract voice features of the dementia patient, and capable of storing dementia information of each subject; a voice output unit for outputting the diagnostic voice data stored in the data memory to the outside so that the subject can hear the diagnostic voice data; a voice receiver for receiving the voice uttered by the subject and converting the voice into voice data; a controller connected to the data memory, the voice output unit, and the voice receiver and controlling the data memory, the voice output unit, and the voice receiver so as to receive voices of the subject after outputting story data, and the controller includes a speech requesting section for requesting the subject to recall contents of the outputted stories and speak by voices after outputting the story data, and includes a voice feature analyzing section for analyzing voice features based on voice data from the subject, and includes a dementia judging section for comparing the voice features of the subject analyzed by the voice feature analyzing section with the patient voice feature data, judging whether dementia is suffered and/or judging a dementia state, and the controller controls such that dementia information of each subject judged by the dementia judging section is stored in the data memory. The story data is formed as a story form centering on a person whose subject is known and an event related to the person, an actual story composed of a plurality of sentences, and is data formed by: extracting a part of words in noun form included in the actual story, and changing the extracted words in noun form to words in other noun form of the same or similar type to form an adapted story, and then converting the adapted story into data of voice data.

Effects of the invention

As described above, the present invention has the following effects: the diagnosis of dementia and the screening of the patients suffering from dementia can be rapidly and easily performed without professional staff, and the early diagnosis of dementia and the screening of the patients suffering from dementia can be rapidly and accurately performed, and the evaluation of questions and answers or the voice prompt of stories can be performed by voice, so that even the elderly with difficulty in recognizing characters can be easily inspected, and hundreds of thousands of patients suffering from dementia in the whole population can be rapidly and simply screened with relatively low cost in the national range.

Drawings

Fig. 1 is a schematic block diagram showing a system for diagnosing dementia using story-based questionnaire type evaluation of the present invention.

Fig. 2 is a schematic block diagram showing a controller of the system for diagnosing dementia using story-based questionnaire type evaluation of the present invention.

Fig. 3 is an exemplary diagram showing an example of a practical story of the system for diagnosing dementia of the present invention.

Fig. 4 is an exemplary diagram showing an example of an adaptation story of the system for diagnosing dementia of the present invention.

Fig. 5 is an exemplary diagram showing an example of story notification text of the system for diagnosing dementia of the present invention.

Fig. 6 is an exemplary diagram showing an example of a problem notification text of the system for diagnosing dementia according to the present invention.

Fig. 7 is an exemplary diagram showing an example of the problem of the system for diagnosing dementia according to the present invention.

Fig. 8 is a schematic block diagram showing a system for diagnosing dementia according to another embodiment of the present invention.

Fig. 9 is a schematic block diagram showing a controller of a system for diagnosing dementia according to another embodiment of the present invention.

Fig. 10 is a step chart showing a method of diagnosing dementia using story-based questionnaire assessment, which uses the present invention.

Fig. 11 is a schematic block diagram showing a system for diagnosing dementia using story-based speech extraction of the present invention.

Fig. 12 is a schematic block diagram showing a controller of the system for diagnosing dementia using story-based speech extraction of the present invention.

Fig. 13 is a step chart showing a method of diagnosing dementia using story-based speech extraction of the present invention.

Detailed Description

The present invention relates to a system for diagnosing dementia using story-based question-answer type assessment, which can rapidly and easily diagnose dementia without a professional, and rapidly and accurately diagnose dementia and screen light patients, and can easily check even elderly people having difficulty in recognizing characters by performing question-answer type assessment through voice. The system for diagnosing dementia comprises: a data inputter 10 capable of inputting an actual story; a data memory 20 for storing text data inputted through the data input unit 10; a voice output unit 30 for outputting voice data for diagnosis to the outside through voice; a voice receiver 40 for receiving an answer voice of the subject; the controller 50 is connected to and controls various devices including the data inputter 10, and forms an adapted story, a question, and a correct answer based on an actual story, and receives an answer of a subject, and compares the answer with the correct answer to judge dementia.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. It should be understood, however, that the invention may be embodied in many different forms and should not be limited to the embodiments described.

Fig. 1 to 7 are diagrams showing a system for diagnosing dementia according to the present invention. As shown in the drawing, the system for diagnosing dementia using story-based questionnaire type assessment according to the present invention includes: a data inputter 10 capable of inputting an actual story; a data memory 20 for storing text data inputted through the data input unit 10; a voice output unit 30 for outputting voice data for diagnosis to the outside through voice; a voice receiver 40 for receiving an answer voice of the subject; the controller 50 is connected to and controls various devices including the data inputter 10, and forms an adapted story, a question, and a correct answer based on an actual story, and cuts out an answer of a received subject, and compares the answer with the correct answer to judge dementia.

The data inputter 10 is a general input means capable of inputting data to a storage device, and allows a user to input text data on an actual story into the data storage 20 through the controller 50.

Here, the actual story is composed of a plurality of sentences as a story form centering on a person whose subject is known and an event related to the person, and is composed and prepared by a producer or an operator.

The actual story as described above is preferably a fairy tale story or a alley story in which the subject is known. For example, the actual story is a fairy tale story of a white snow princess as shown in fig. 3 or a moral story as an miso story, the plot of which can be summarized within 15 sentences.

Here, the subject is known to mean that the subject has already known a long time ago for classical, conventional or well known fairy tales, allelochemicals, and the like, and can easily recall these episodes.

Preferably, a plurality of actual stories as described above are prepared so that various fairy tale stories or alley stories can be used in order to use different stories when the same subject is repeatedly checked, thereby improving accuracy, while the fairy tale stories or alley stories known to the subject can be used in diagnosis by appropriately selecting them.

The data memory 20 is a storage device (e.g., a memory or a hard disk) that can store digital data, in which various data including text data input through the data input unit 10 are stored, and dementia information of each subject can be stored.

The voice outputter 30 may output digital data through a voice output device (e.g., a speaker) and may output diagnostic voice data to the outside so that the subject can hear. Here, the diagnostic voice data refers to voice data of an adaptation story and a question outputted for determining dementia.

The speech receiver 40 is a device (e.g., a microphone) that receives speech for receiving answers spoken by a subject.

The controller 50 is connected to the data input unit 10, the data storage 20, the voice output unit 30 and the voice receiver 40, and the controller 50 includes: a story data formation section 51 that automatically forms adapted story data, question data, and correct answer data based on an actual story input in the form of text data; an answer data forming unit 52 for receiving an answer from the subject and forming answer data; the dementia judging section 53 compares and analyzes the correct answer data and the answer data to confirm a wrong answer rate, and judges whether or not a dementia is present and/or judges a dementia state based on the degree of the wrong answer rate.

The controller 50 as described above is configured to control the data memory 20, the voice outputter 30, and the voice receiver 40 so as to sequentially output the story-adapting data and the question data, and to receive answers of the subjects to the output questions, thereby storing the judged dementia information of each subject.

The story data formation section 51 is for automatically forming adapted story data, question data, and correct answer data based on text data of an input actual story.

Here, the adaptation story data is data formed by: extracting a part of the words in the noun form included in the actual story, changing the extracted words in the noun form into words in other noun forms of the same or similar type to form an adapted story, and converting the adapted story into voice data.

That is, the adaptation story is formed by: extracting a part of words in the form of nouns included in the actual story, and changing the extracted words in the form of nouns to other words of a similar type.

As an example, as shown in fig. 3, words in the form of nouns extracted from an actual story are marked with red boxes, and words in the form of nouns extracted marked with red boxes in fig. 3 are "white snow princess", "relay mother", "seven", "short person", "apple", "horse", and "prince".

As shown in fig. 4, words in the form of nouns extracted in the actual story as described above may be changed to other words of similar types, for example, "black and snow princess", "mastery", "five", "miner", "peach", "donkey", and "miner".

That is, the "white snow princess" of the actual story is changed to the "black snow princess", the "succession mother" of the actual story is changed to the "aunt", the "seven" of the actual story is changed to the "five", the "short person" of the actual story is changed to the "miner", the "apple" of the actual story is changed to the "peach", the "horse" of the actual story is changed to the "donkey", and the "prince" of the actual story is changed to the "miner".

As described above, the adapting story shown in fig. 4 is completed by changing a part of words in noun form of the actual story shown in fig. 3.

The adaptation as described above consists of 5 to 15 sentences, preferably comprising 20 to 50 words of different noun form, because if the number of sentences is less than 5 or the number of words of noun form is comprised less than 20, the length of the whole adaptation story is too short to be remembered easily and thus the short-term memory of the subject cannot be tested effectively or the dementia patient cannot be screened appropriately.

In addition, if the number of sentences of the adaptation story is more than 15 or words in the form of nouns are included more than 50, the adaptation story itself becomes too long to be remembered, and thus there is a risk of misclassifying a normal person as a dementia patient.

The adapted story preferably includes at least one additional sentence made by an individual, the additional sentence being different from the content of the actual story and being related and linked to the content of the adapted story.

An example of the additional sentence described above is shown by a portion shown in a of fig. 4, and functions to interfere with memory by causing confusion and confusion on the subject. That is, the subject cannot well remember the overall contents of the adapted story due to the memory disturbance of the additional sentence, especially in the case of the dementia patient, thereby screening the dementia patient more accurately.

Here, the additional sentence is different from the content of the actual story, meaning that it contains the content not in the content of the actual story, and the content of the additional sentence related to and connected with the content of the adapted story, meaning that it is connected with the story development of the adapted story. Such additional sentences may be personally produced by the producer and input through the data input unit 10, or may include contents which are public to the owner and cause the stimulus, automatically formed by the controller 50.

In addition, as shown in fig. 4, the additional sentence is preferably placed at the end of the adaptation story in order to effectively interfere with the memory of the adaptation story by the subject.

The problem data is data formed by: and taking words in the modified noun form included in the adaptation story as correct answers, forming a plurality of questions related to the adaptation story, and converting the questions into voice data.

Here, the problem is preferably composed of 3 to 12 because when the number of problems is less than 3, the memory of the subject may not be properly tested due to low discrimination, and when the number of problems is more than 12, the accurate memory test may not be performed due to exceeding the short-term memory of the subject.

In addition, the question is preferably a question regarding whether the event, behavior, scenario and content presented in the adapted story is realistic or not, in order to conduct an appropriate short-term memory test by means of a simple and clear fact-based question.

In addition, as shown in fig. 7, the question is preferably in the form of selecting one of a plurality of options, wherein the wrong answer includes words in the form of nouns included in the actual story of the same or similar type as the correct answer. For example, in the case where the question shown in fig. 7 is in the form of selecting one from four options, words in the form of nouns included in the actual story, that is, "white snow princess", "relay mother", "seven", "short person", "apple", "horse", and "prince" are included in the 4 options as wrong answers, causing the subject to memorize confusedly, thus causing the subject to select the known fairy tale content as wrong answer, so that it can be appropriately judged as a dementia patient.

That is, some of the subjects do not remember too much the content of the story-adapting including the correct answer to the question, but rather remember the content of the actual story remembered long ago and select it as an answer, resulting in a wrong answer, and the subject producing more of such wrong answer is diagnosed and screened as a dementia patient.

In addition, a question confirming the correct answer rate of the subject itself is preferably configured as the last question, and preferably, is arranged. This problem of confirming the correct answer rate of itself is shown in part D of fig. 7, allowing the operator to evaluate the meta-awareness ability of the subject to recognize the content known and not known by himself. This meta-cognitive assessment can be used as data to once again confirm the accuracy of the diagnosis of dementia by the system.

The correct answer data is data formed by: extracting a correct answer corresponding to the question, and converting the extracted correct answer into voice data. The correct answer data as described above is used for comparing with the answer data collected from the subject and confirming the knowledge of the wrong answer rate.

Here, the correct answer refers to a word in the form of a noun which is changed in the adaptation story after being extracted from the actual story, and the adaptation story of fig. 4 is taken as an example, and the "black snow princess", "ault mother", "five", "miner", "peach", "donkey" and "miner" marked with red boxes belong to the correct answer.

The answer data forming unit 52 is configured to receive an answer from a subject by voice and automatically form answer data. Here, the answer data is voice data of an answer to the question by the subject received and stored by the voice receiver 40.

The dementia judging unit 53 compares the generated correct answer data with the answer data, and judges whether or not the patient has dementia and/or judges the state of dementia by comparing the incorrect answer rate of the questions.

That is, the problem of the subject to be examined about the adapted story "snow white" is confused with the actual story of the known "snow white", and if there are many wrong answers so initiated, the wrong answer rate becomes high, and the subject is judged as a dementia patient.

The dementia judging section 53 confirms a correct answer and a wrong answer by comparing and analyzing the voice of the correct answer data and the voice of the answer data in order to make a more accurate wrong answer rate judgment by voice recognition and voice comparison.

Here, the correct answer to the question is marked with a red box in fig. 7, and of course, options other than the correct answer are regarded as wrong answers, and in fig. 7, the correct answer to the "question 8" is confirmed by evaluating the "question 1" to the "question 7", and the wrong answer rate is the ratio of the wrong answer to all questions.

Preferably, in the dementia judging section 53 as described above, if the wrong answer rate is higher than at least 25%, dementia is judged, if the wrong answer rate is 25% to 40%, mild dementia is judged, and if the wrong answer rate is higher than at least 40%, severe dementia is judged, which is obtained by performing a clinical experiment on a plurality of subjects, and comparing, analyzing and confirming the wrong answer rates of the subjects belonging to the normal group and the subjects belonging to the dementia group.

That is, if the wrong answer rate is higher than 25%, all dementia patients are confirmed, and between 25% and 40%, mild dementia patients are distinguished, and if it is higher than 40%, severe dementia patients are distinguished.

The controller 50 receives a story notification text for notifying an output of the adapted story while notifying a problem that is presented in relation to the adapted story after the output of the adapted story, and converts the story notification text into voice data, and the controller 50 controls the voice outputter 30 to output the story notification text first before outputting the adapted story.

The story notification text is created and input in the form of text data by a producer or operator, and then converted into voice data and output.

An example of the story notification text described above is shown in fig. 5, and the output of the story notification text is used to prompt the subject in advance, that is, to output the adapted story in a voice form, and then to present a question, and to answer the question.

The story notification text preferably places a sentence at the beginning that includes the title of the fairy tale story or the alley story presented as the adapted story. The title of the alleged story or fairy tale as described above is, as shown by section B of fig. 5 ("white princess"), the subject is caused to recall in advance the relevant fairy tale "white princess" by the title of the fairy tale (e.g., "white princess") output by voice prompt in the story notification text, so that the entire plot of "white princess" can be easily remembered.

The controller 50 receives a question notification text for notifying that a predetermined number of questions are to be posed, converts the question notification text into voice data, and controls the voice outputter 30 to output the question notification text after the output of the adapted story is completed.

The problem notification text is created and input in the form of text data by a creator or an operator, and then converted into voice data and output.

The output of the question notification text as described above is used to notify that questions are output immediately after the story is adapted to be output and the questions need to be answered, while the number of questions is presented in advance as shown by section C (7 questions) of fig. 6 so that the subject appropriately handles the output questions.

The actual stories of the system for diagnosing dementia are prepared and respectively input and formed into a plurality of adapted stories, and questions and correct answers associated with each of the adapted stories are formed in order to perform various questionnaire-type evaluations through the plurality of actual stories, thereby more accurately performing diagnosis and judgment of dementia.

As described above, in a state where a plurality of actual stories are input, the controller 50 further includes: a known story confirmation unit 54 that outputs a story confirmation text by voice to a subject to confirm which of a plurality of adapted stories are the known stories, before outputting the adapted event; the story selection section 55 selects an adapted story which is confirmed to be known to the subject to be examined, so as to output the adapted story confirmed by the known story confirmation section 54.

For example, the known story confirmation section 54 outputs story confirmation text in which titles of a plurality of adapted stories (for example, "white snow princess", "gray girl", "herring princess", "pinacoh", and "prince and beggar") are recorded by voice so that the subject can confirm which fairy tale stories or alley stories are among the adapted stories stored in the diagnostic device.

The story selection section 55 allows the subject to listen to the story confirmation text and then select a known adapted story. For example, the adapted stories of "white snow princess", "gray girl", "herring princess", "pinacoh", and "prince and a gas" are stored, and the titles of these stories are output, in which state if the subject knows "gray girl", the "gray girl" is selected by voice answer or by button operation.

Thus, if "calyx" is selected, in order to diagnose dementia, an adaptation story and a question related to "calyx" are output to the subject, so that diagnosis of dementia is performed with the known fairy tale story or the alley story of the subject as a standard, while appropriately preventing diagnosis of dementia based on the unknown fairy tale story or the alley story of the subject, thereby improving accuracy of diagnosis of dementia.

The controller 50 further includes: using the story confirmation section 56, before outputting the adaptation event, a voice is output to the subject to confirm which of the plurality of adaptation stories was previously used for the diagnostic check; the story removing section 57 removes the adapted story confirmed as previously used by the used story confirming section 56 from the output.

The used story confirming section 56 is for confirming a story previously used in a diagnostic check of the subject himself/herself among a plurality of adapted stories, and the story excluding section 57 is for excluding and excluding the adapted stories confirmed as being used so as not to be outputted to the subject.

As described above, by excluding the adaptation story that the subject has used in the diagnostic check, it is possible to appropriately prevent the degradation of the accuracy of diagnosis due to the repeated use of the same content and the same problem.

The dementia diagnosis system constructed as described above may be constituted by an integrated dementia diagnosis device, which is formed by the data input unit 10, the data memory 20, the voice output unit 30, the voice receiver 40 and the controller 50 being modularized and integrated, or by an automatic dementia diagnosis device, which automates all functions by programming, artificial intelligence, or the like.

As described above, the system for diagnosing dementia is a breakthrough invention that changes the main words of the classical actual story, which are known and familiar to the subject, into new words and presents them as an adapted story, thereby bringing about the consolidation of memory, while maximizing the disturbing effect of memory by presenting the original words of the classical actual story as wrong answers, thus enabling rapid and accurate screening of dementia.

Fig. 8 to 9 are schematic views showing a system for diagnosing dementia according to another embodiment of the present invention. As shown, the system for diagnosing dementia according to another embodiment of the present invention further includes: a display 60 connected to the controller 50 and capable of displaying text data on a screen; a text input 70 connected to the controller 50 and capable of modifying text data by inputting text.

The display 60 is a device (e.g., monitor) capable of displaying text or images on a screen for displaying the adaptation data and the question data in the form of text data.

The text inputter 70 is a device (e.g., a keyboard) capable of inputting and modifying text, and is used for text data capable of modifying the adaptation data and the question data.

The controller 50 further includes: a story confirmation and modification unit 58 for displaying the adapted story in the form of text data on the display 60 so that a producer or operator can confirm the adapted story and perform manual modification by the text input unit 70; and a question confirmation and modification unit 59 for displaying the question on the display 60 in the form of text data so that a producer or operator can confirm the question and manually modify the question by the text input unit 70.

The story confirmation modification section 58 goes through a process of personally confirming and modifying the adapted story automatically adapted by the controller 50 by a producer or an operator or the like so as to form a more natural and smooth high-quality adapted story.

The problem confirmation modification unit 59 confirms and modifies the process of the problem automatically formed by the controller 50 by a producer or an operator or the like in person so that a problem of natural and smooth difficulty of adjustment can be formed.

The system for diagnosing dementia further comprises a data transmitter 80 connected to the data storage 20 and controlled by the controller 50. For example, the data transmitter 80 is a wired and wireless transceiver device, which is controlled by the controller 50 and transmits the dementia information of each subject stored in the data memory 20 to an external device through a transmission line or network.

Fig. 10 is a step chart showing a method of diagnosing dementia according to the present invention.

Referring to fig. 10, the method for diagnosing dementia using story-based questionnaire assessment according to the present invention includes the steps of: an actual story preparation step S1 of preparing an actual story; an actual story input step S2 of converting the actual story into text data and inputting the text data to a diagnostic device; an adapted story forming step S3 of forming an adapted story by adapting the inputted actual story; a question forming step S4 of forming a plurality of questions having words in the form of altered nouns included in the adaptation story as correct answers; a voice data conversion step S5 of converting the adapted stories and questions into voice data; an adapted story output step S6 of outputting the voice data of the adapted story to the outside; a question outputting step S7 of sequentially outputting the voice data of the question to the outside; an answer receiving step S8 of receiving an answer to the question from the subject; and a dementia judging step S9, judging whether the dementia is caused according to the wrong answer rate obtained by the questions and the answers. Dementia is diagnosed by sequentially performing the steps as described above.

In the question forming step S4, the question is composed of 3 to 12 questions, which are questions related to whether or not the event, behavior, scenario, and content shown in the adapted story are real, and the question is in the form of selecting one of a plurality of options, wherein a word in the form of noun included in the actual story of the same or similar type as a correct answer is included in a wrong answer, and a question confirming the correct answer rate of the subject himself is configured as a final question.

In the actual story preparation step S1, the actual story may be a fairy tale story or a alley story in which the subject is known.

In the story-adapting forming step S3, the story-adapting matter is composed of 5 to 15 sentences and includes 20 to 50 words in different noun forms, the story-adapting forming step S3 further includes: a sentence attachment process of attaching at least one artificial sentence which is different from the content of the actual story and which is related and connected to the content of the adapted story, the additional sentence being placed at the end of the adapted story.

The step S6 of adapting a story output further includes: a story notification output process of, before outputting the adaptation event, notifying, by voice, the output of the adaptation story, and simultaneously notifying, by voice, a story notification text for showing a problem related to the adaptation story after outputting the adaptation story, the adaptation story being a fairy tale story or a story in which a subject is known, the story notification output process further including a process of outputting, to a front of the story notification text, a sentence including a title of the fairy tale story or the story in which the adaptation story is shown.

The adapting story output step S6 may further include: and a question notification output process of outputting a question notification text for notifying that a predetermined number of questions are to be posed by voice after the output of the adapted story is completed.

In the dementia judging step S9, if the wrong answer rate is higher than at least 25%, dementia may be judged, if the wrong answer rate is 25% to 40%, mild dementia may be judged, and if the wrong answer rate is higher than at least 40%, severe dementia may be judged.

Preparing and respectively inputting a plurality of said actual stories and forming a plurality of adapted stories, and forming questions associated with each of said adapted stories, said adapted story outputting step S6 further comprising: a cognitive story confirmation process of outputting story confirmation text to the subject by voice to confirm which of a plurality of adapted stories are known, before outputting the adapted event; and a story selection process of selecting an adaptation story which is confirmed to be known to the subject to be examined, so as to output the adaptation story which is confirmed by the cognitive story confirmation process.

Preparing and respectively inputting a plurality of the actual stories and forming a plurality of adapted stories, and forming a question associated with each of the adapted stories, the adapted story outputting step S6 may further include: using a story confirmation process, before outputting the adaptation event, outputting to the subject a confirmation of which of the plurality of adaptation stories the previously used diagnostic check stories were; and a story elimination process of eliminating from the output the adapted story confirmed as previously used by the use story confirmation process.

The adapting story forming step S3 may further include: and a story confirmation and modification process for displaying the adapted story formed by the text data on a display so that a producer or operator can confirm and manually modify the adapted story.

The problem forming step S4 may further include: and a question confirmation and modification process for displaying the question formed by the text data on a display so that a producer or an operator can confirm and manually modify the question.

The dementia information of each subject may be transmitted by the data transmitter 80 as described above, so that the information of the dementia patients diagnosed and screened may be easily provided to the management server of the country and private authorities for dementia management and screening.

Fig. 11 to 12 are diagrams showing the system for diagnosing dementia of the present invention. As shown in the drawing, the system for diagnosing dementia using story-based speech extraction according to the present invention includes: a data memory 100 for storing diagnosis data for diagnosing dementia and patient voice feature data; a voice output unit 200 for outputting voice data for diagnosis stored in the data storage 100 to the outside by voice; a voice receiver 300 that receives a voice of a subject; and a controller 400 connected to and controlling the data storage 100, the voice output unit 200, and the voice receiver 300, and judging dementia by voice characteristics of a subject and voice characteristic data of a patient.

The data storage 100 is a device (e.g., a memory or a hard disk) that can store digital data, and stores voice data for diagnosis including story data for diagnosing dementia, and stores patient voice feature data, wherein voice features of a dementia patient are extracted by comparing and analyzing voice features of a normal person and voice features of a dementia patient, and dementia information of each subject can be stored.

Here, the patient voice feature data is data obtained by differentiating from voice features of a normal person and digitizing common voice features extracted from voices of a plurality of dementia patients.

The voice characteristics of the normal person and the dementia patient comprise any one or more than two of basic frequency, speech speed, amplitude perturbation, fundamental frequency perturbation, formants and voice frequency spectrum of voice.

The fundamental frequency of speech refers to the frequency representing the average, habitual pitch that speech has, and the speech speed refers to the speaking speed and the pronunciation speed, representing the speed at which a person speaks.

The amplitude perturbation (shimmer 0) numerically represents a regular change in the amplitude change of the voice, the fundamental frequency perturbation (jitter 0) numerically represents the rate of change of the vocal cord vibration, the formants (formant 0) represent frequency components featuring a set, and the voice spectrum represents the distribution of each frequency of the voice.

In addition, the story data forms an actual story composed of a plurality of sentences as a story form centering on a person whose subject is known and an event related to the person, and is data formed by: extracting a part of the words in the noun form included in the actual story, and then changing the extracted words in the noun form into words in other noun forms of the same or similar type to form an adapted story and converting the adapted story into voice data.

The story data is data formed by: the adapted story is converted into voice data so that the adapted story can be easily output by voice and provided to the subject at a later time.

The actual story is preferably a fairy tale story or a alley story for which the subject is known. For example, the actual story is a fairy tale story of a white snow princess as shown in fig. 3 or a moral story as an eastern story, and a plot summarized as 15 sentences or less may be used.

Here, the subject is known to mean that the subject has already known a long time ago for classical, conventional or well-known fairy tales, allelochemicals, and the like, and can easily recall these episodes.

The adapting story is to adapt the actual story, i.e. to extract a part of the words in the form of nouns comprised in the actual story and to change the extracted words in the form of nouns to other words of a similar type. As an example, as shown in fig. 3, words in the form of nouns extracted from an actual story are marked with red boxes. The words in the form of extracted nouns marked with red boxes are "white snow princess", "relay mother", "seven", "short person", "apple", "horse" and "prince".

When changing to the similar words described above, the words may be changed automatically by extracting the similar words, or directly by the producer or operator participating in person. Here, it is preferable that the word is changed personally by a producer or an operator or a part of the word is intervened so that the adaptation becomes natural.

The adaptation story shown in fig. 4 is completed by the change of a word in the form of a part of nouns shown in the actual story as described above.

Here, preferably, when the number of sentences of the actual story exceeds 15, in the process of adapting it to an adapted story, intervention is made by a producer or operator to reduce the number of sentences, or preparation is made so that the actual story itself has a number of sentences within 15.

In addition, preferably, the adapted story includes at least one additional sentence made by human, the additional sentence being different from the content of the actual story and being related and connected to the content of the adapted story.

An example of the additional sentence as described above is shown by part a of fig. 4, which is used to confuse and confuse the subject to interfere with memory. That is, the subject cannot well remember the overall contents of the adapted story due to the memory disturbance of the additional sentence, especially in the case of the dementia patient, thereby screening the dementia patient more accurately.

Here, the additional sentence is different from the content of the actual story, meaning that it contains the content not in the content of the actual story, and the additional sentence is the content related to and connected with the content of the adapted story, meaning that it is the content connected with the story development of the adapted story. Such additional sentences may be personally produced by the producer or included to include content that is public to the host and causes the stimulus, and automatically formed and produced.

In addition, as shown in fig. 4, the additional sentence is preferably located at the end of the adaptation story in order to effectively interfere with the memory of the adaptation story of the subject.

The adapted story completed in this way is converted into voice data and stored in the data storage 100, and then can be easily outputted and provided to a subject by voice through the voice outputter 200 after a day.

The voice data conversion as described above refers to voice data in which a story is adapted by human reading and then converted into digital data by various digital voice collecting devices, recorders, and the like, or text is directly converted into mechanical sound by a text-to-speech converting device.

In addition, the data storage 100 may store question data and correct answer data regarding the adapted story.

The problem data is data formed by: and taking words in the modified noun form included in the adaptation story as correct answers, forming a plurality of questions related to the adaptation story, and converting the questions into voice data. The question data is used to test the memory of the subject and trigger the utterance.

The correct answer data is data of a correct answer to the question including words in the form of altered nouns included in the adaptation story.

The problems are preferably composed of 3 to 12 because when the number of problems is less than 3, the memory of the subject may not be properly tested because the content spoken by the subject is reduced and the discrimination is low, and when the number of problems is more than 12, the accurate memory test may not be performed because the short-term memory of the subject is exceeded.

In addition, the question is preferably a question regarding whether or not the event, behavior, scenario, and content shown in the adaptation story are realistic, in order to perform an appropriate short-term memory test based on a simple and clear fact, and to easily induce a subject to speak.

The question may be a question that is requested to be answered with a sentence or word in order to induce the subject to answer in a short text or short manner, thereby ensuring that more speech is acquired while dementia judgment is made, while extracting speech features more accurately.

In addition, as shown in fig. 7, the question may be in the form of selecting one option from a plurality of options, wherein the wrong answer includes words in the form of nouns included in the actual story of the same or similar type as the correct answer. For example, in the case where the question shown in fig. 7 is in the form of selecting one from four options, words in the form of nouns included in the actual story, that is, "white snow princess", "relay mother", "seven", "short person", "apple", "horse", and "prince" are included in the 4 options as wrong answers, causing the subject to memorize confusedly, thus causing the subject to select the known fairy tale content as wrong answer, so that it can be appropriately determined as a dementia patient.

That is, some of the subjects do not remember too much the content of the story-adapting including the correct answer to the question, but rather remember the content of the actual story remembered long ago and select it as an answer, resulting in the speech features and wrong answer of the dementia patient, and the subject who produces more of such speech features and wrong answer of the dementia patient is diagnosed and screened as the dementia patient.

In addition, it is preferable to configure a question for confirming the correct answer rate of the subject itself as the last question. This problem of confirming the correct answer rate of itself, as shown by part D of fig. 7, allows the operator to evaluate the meta-awareness ability of the subject to recognize the content that is known and unknown to himself. This meta-cognitive assessment can be used as data to once again confirm the accuracy of the diagnosis of dementia by this method.

The voice outputter 200 may be a device (e.g., a speaker) that outputs digital data by voice, and functions to output the diagnostic voice data stored in the data memory 100 to the outside so that the subject can hear.

When the subject listens to the "snow white" story of fig. 4 through the voice output unit 200, the subject is confused with the "snow white" story of fig. 3, which is known and memorized by the subject itself, that is, the memory disturbance occurs.

As described above, in a state where the subject has memory disturbance, recall of the content of the story adaptation output to the subject is requested and spoken by voice, thereby generating an environment where the memory of the subject is degraded or confusion is caused, and it is possible to rapidly and accurately diagnose and screen the subject suffering from dementia, which is difficult to easily overcome such an environment.

In addition, by performing a diagnostic test by adapting the speech output of the story, an elderly subject who has little or no knowledge of the text can also be easily screened for diagnostic dementia.

The voice receiver 300 is a device (e.g., a microphone) that can receive voice and convert it into digital data, and is used to receive voice uttered by a subject and convert it into voice data.

That is, after the voice of the subject is received by the voice receiver 300, the voice characteristics thereof are analyzed by the controller 400 and then compared with the voice characteristic data of the patient, thereby being used for dementia judgment.

The controller 400 is connected to the data storage 100, the voice outputter 200, and the voice receiver 300, and controls the data storage 100, the voice outputter 200, and the voice receiver 300 so as to receive voices of the subject after outputting story data.

In addition, the controller 400 includes a speech requesting part 410 that requests a subject to recall and speak the content of the output story after the output of the story data is completed, and includes a voice feature analyzing part 420 that analyzes voice features based on voice data from the subject, and includes a dementia judging part 430 that compares the voice features of the subject analyzed by the voice feature analyzing part 420 with the patient voice feature data to judge whether dementia is present and/or to judge a dementia state.

The controller 400 as described above is configured to analyze the voice characteristics of the subject and perform dementia judgment based on the analyzed voice characteristics, so as to control the storage of the dementia information of each subject judged by the dementia judgment part 430 in the data memory 100.

The speech requesting part 410 is used for requesting the checked object to recall and speak the content of the adapted story, and the speech requesting part 410 prompts the time of the checked object to speak, and simultaneously promotes and induces the speech. For example, the speech requesting section 410 as described above requests the content by speech output such as "recall and speak the content of the story".

The voice characteristic analysis part 420 is used for analyzing the voice of the detected object collected by the voice receiver 300 and extracting the voice characteristic of the detected object. That is, the voice characteristic analysis unit 420 is configured to analyze and extract a fundamental frequency, a speech velocity, an amplitude perturbation, a fundamental frequency perturbation, a formant, and a voice spectrum of a voice from voice data of a subject.

The dementia judging section 430 is for judging dementia of a subject by comparing the voice feature of the subject with the patient voice feature data which is the voice feature of the dementia patient. Here, when the subject recalls and speaks the "white snow princess" of the adapted story, which is adapted and presented, the subject is confused with the actual story of the known "white snow princess", and the voice characteristic of the dementia patient is caused, and at this time, it is judged as dementia.

The controller 400 further includes: the answer requesting unit 440, after completing the output of the adaptation event to the subject, sequentially outputs the voice data of the question to the outside and requests the subject to answer the question. The answer requesting unit 440 is used to confirm the memory of the subject by a question and to induce a speech by an answer.

The controller 400 further includes: the auxiliary dementia judging unit 450 confirms a wrong answer rate by comparing and analyzing answers collected from the subject for the question with the right answer, and judges whether or not the dementia is present and/or judges the dementia state based on the degree of the wrong answer rate. The auxiliary dementia judging unit 450 is for assisting in judging dementia by the wrong answer rate of the answer to the question by the subject.

Here, the correct answer to the question is marked with a red box in fig. 7, and of course, options other than the correct answer are regarded as wrong answers. In fig. 7, the correct answer to "question 8" is confirmed by evaluating "question 1" to "question 7", and the wrong answer rate is the ratio of the wrong answer to all questions.

Preferably, the controller 400 as described above judges dementia of a subject by comprehensively performing dementia judgment according to the voice characteristics of the subject and dementia judgment according to the wrong answer rate of the subject, in order to improve the accuracy of dementia judgment.

The controller 400 further includes: the memory disturbing section 460, before completing the output of the adapted story and requesting the subject to speak by voice, makes a period of time pass to disturb the memory of the subject. The memory disturbing part 460 disturbs the memory of the subject for the story adaptation by time delay so that the subject has difficulty in completely remembering the contents of the story adaptation, thereby screening out the dementia patient more accurately.

At this time, the time elapsed for the period of time may be several seconds to several tens of seconds, and audio such as music is output during the period of time, so that the memory of the subject may be more positively disturbed.

The dementia diagnosis system constructed as described above may be constituted by an integrated dementia diagnosis device, which is modularized and integrally formed by the data memory 100, the voice output unit 200, the voice receiver 300, and the controller 400, or may be constituted by an automatic dementia diagnosis device, which automates all functional programming, artificial intelligence, or the like.

The system for diagnosing dementia further comprises: a data inputter 500 connected to the data storage 100 and controlled by the controller 400; and a data transmitter 600 connected to the data storage 100 and controlled by the controller 400.

The data inputter 500 may be controlled by the controller 400 and input and store various data including voice data for diagnosis into the data memory 100. That is, the operator can easily input various data including the voice data for diagnosis into the system, so that the change, addition, and update of the voice data for diagnosis are appropriately performed.

The data transmitter 600 is a wired and wireless transceiver controlled by the controller 400 and transmits the dementia information of each subject stored in the data memory 100 to an external device through a transmission line or network.

Since the dementia information of each subject can be transmitted through the data transmitter 600 as described above, the information of the dementia patients diagnosed and screened can be easily provided to the management server of the country and private authorities for dementia management and screening.

The system for diagnosing dementia constructed as described above is a breakthrough invention that changes the main words of the classical actual story, which are known and familiar to the subject, into new words and presents them as an adaptation story, bringing about the consolidation of memory while recall and speaking the adaptation story, thereby maximally enhancing the interference effect between the adaptation story and the classical actual story, and thus enabling rapid and accurate screening of dementia.

A system for diagnosing dementia using story-based speech extraction, comprising: a data memory 10 that stores voice data for diagnosis including story data for diagnosing dementia, and stores patient voice feature data that is capable of storing dementia information of each subject by extracting voice features of a dementia patient by comparing and analyzing voice features of a normal person and voice features of the dementia patient; a voice output unit 20 for outputting the diagnostic voice data stored in the data memory 10 to the outside so that the subject can hear the data; a voice receiver 30 that receives a voice uttered by the subject and converts the voice into voice data; a controller 40 connected to the data memory 10, the voice outputter 20, and the voice receiver 30, and controlling the data memory 10, the voice outputter 20, and the voice receiver 30 so as to receive voices of the subject after outputting story data, and including a speech requesting section 41 requesting the subject to recall and speak contents of the output story after completing the outputting of the story data, and including a voice feature analyzing section 42 analyzing voice features based on the voice data from the subject, and including a dementia judging section 43 comparing the voice features of the subject analyzed by the voice feature analyzing section 42 with the patient voice feature data to judge whether there is dementia and/or judge a dementia state. The story data is data formed by forming an actual story composed of a plurality of sentences as a story form centering on a person whose subject is known and an event related to the person, the story data being data formed by: extracting a part of words in noun form included in the actual story, and changing the extracted words in noun form to words in other noun form of the same or similar type to form an adapted story, and then converting the adapted story into data of voice data.

The data storage 10 stores question data and correct answer data of the adapted story, the question data having words in the form of altered nouns included in the adapted story as correct answers and forming a plurality of questions related to the adapted story, the questions being data converted into voice data, the controller 40 further comprising: after completing the output of the adaptation to the subject, the answer requesting unit 44 sequentially outputs the speech data of the questions to the outside, and requests the subject to answer the questions.

A system for diagnosing dementia using story-based speech extraction, wherein the questions consist of 3 to 12 questions about whether the events, behaviors, situations and contents shown in the adapted story are real or not.

A system for diagnosing dementia using story-based speech extraction, wherein the question is a question that is requested to be answered with a sentence or word.

A system for diagnosing dementia using story-based speech extraction, wherein the question is in the form of selecting one option from a plurality of options, wherein a wrong answer includes words in the form of nouns included in the actual story of the same or similar type as a correct answer, and a question confirming a correct answer rate of a subject himself is configured as a last question.

A system for diagnosing dementia using story-based speech extraction, wherein the controller 40 further comprises: the auxiliary dementia judging unit 45 confirms a wrong answer rate by comparing and analyzing answers collected from the subject for the question with the right answer, and judges whether or not the patient has dementia and/or judges the dementia state based on the degree of the wrong answer rate.

A system for diagnosing dementia using story-based speech extraction, wherein the controller 40 determines dementia of a subject by comprehensively performing dementia determination according to speech features of the subject and dementia determination according to wrong answer rates of the subject.

A system for diagnosing dementia using story-based speech extraction, wherein the actual story is a fairy tale story or a moral story known to a subject.

A system for diagnosing dementia using story-based speech extraction, wherein the adaptation events consist of 5 to 15 sentences and include words of 20 to 50 different noun forms.

A system for diagnosing dementia using story-based speech extraction, wherein the adapted story includes at least one additional sentence made by human, which is different from the content of the actual story and related and connected to the content of the adapted story and placed at the end of the adapted story.

A system for diagnosing dementia using story-based speech extraction, wherein the controller 40 further comprises: the memory disturbing section 46 makes a period of time pass to disturb the memory of the subject before the output of the adapted story is completed and the subject is requested to speak by voice.

A system for diagnosing dementia using story-based speech extraction, wherein the speech features are at least one or more of fundamental frequency, speech speed, amplitude perturbation, fundamental frequency perturbation, formants, and speech spectrum of speech.

A system for diagnosing dementia using story-based speech extraction, wherein the system for diagnosing dementia further comprises: a data inputter 50 connected to the data memory 10 and controlled by the controller 40, and operable to input and store various data including diagnostic voice data into the data memory 10; a data transmitter 60 connected to the data memory 10 and controlled by the controller 40, and may transmit the dementia information of each subject stored in the data memory 10 to an external device.

Fig. 13 is a step chart showing a method of diagnosing dementia by story-based speech extraction of the present invention.

Referring to fig. 13, the method for diagnosing dementia using story-based speech extraction according to the present invention includes the steps of: a patient data preparing step S10 of extracting the voice characteristics of the dementia patient and preparing them as the patient voice characteristic data; an actual story preparing step S20 of preparing an actual story; an adapted story forming step S30 of preparing an adapted story by adapting the actual story and converting it into voice data; an adapted story output step S40 of outputting voice data of the adapted story to the outside; a speech requesting step S50 of requesting the subject to speak the content of the adapted story; a voice collection step S60 for receiving and collecting voices uttered by the subject; and a dementia judging step S70 of judging whether or not the patient suffers from dementia by comparing the collected voice of the subject with the patient voice characteristic data. Dementia is diagnosed by sequentially performing the steps as described above.

The method of diagnosing dementia may further include: a question forming step S30a of preparing a plurality of questions related to the adaptation story with words in the form of altered nouns included in the adaptation story as correct answers, and converting the plurality of questions into voice data after the adaptation story forming step S30. The step S50 of requesting the speech may further include: and a question output process of outputting the voice data of the question to the outside in sequence after finishing the output of the story adaptation to the detected object, and raising the question to the detected object.

In the question forming step S30a, the question is composed of 3 to 12, and the question may be a question regarding whether the event, the behavior, the scenario, and the content shown in the adaptation story are real.

The question may be a question that is requested to be answered with a sentence or word.

The question is in the form of selecting one option from a plurality of options, wherein the wrong answer includes words in the form of nouns included in the actual story of the same or similar type as the correct answer, and a question confirming the correct answer rate of the subject itself is configured as the last question.

The dementia judging step S70 may further include: and assisting a dementia judging process of confirming a wrong answer rate by comparing and analyzing answers collected from the subject for the question with correct answers, and judging whether dementia is suffered from and/or judging a dementia state according to the degree of the wrong answer rate.

The dementia judging step S70 can judge dementia of the subject by performing dementia judgment according to the voice characteristics of the subject and dementia judgment according to the wrong answer rate of the subject in combination.

In the actual story preparation step S20, the actual story may be a fairy tale story or a alley story in which the subject is known.

In the adaptation story forming step S30, the adaptation event is composed of 5 to 15 sentences, and may include words in the form of 20 to 50 different nouns.

The adapting story forming step S30 further includes: a sentence attaching process of attaching at least one artificial sentence which is different from the content of the actual story and which is related and connected to the content of the adapted story, the additional sentence being placed at the end of the adapted story.

The step S50 of requesting the speech may further include: a memory disruption process of causing a period of time to elapse to disrupt the memory of the subject before completing the outputting of the adapted story and requesting the subject to speak by speech.

In the patient data preparing step S10, the voice feature may be at least one or two or more of a fundamental frequency, a speech velocity, an amplitude perturbation, a fundamental frequency perturbation, a formant, and a voice spectrum of voice.

Embodiments according to the present invention may be implemented in the form of a computer program executable by various components on a computer, and such a computer program may be recorded on a computer-readable medium. At this time, the medium may include magnetic media such as hard disks, floppy disks, and magnetic disks, optical recording media such as CD-ROMs, and DVDs, magneto-optical media (magneto-optical media) such as floppy disks (floppy disks), and hardware devices such as ROMs, RAMs, flash memories, etc., which store and execute program instructions.

On the other hand, the computer program may be specially designed and configured for the present invention, or may be known and available to those skilled in the computer software arts. As an example of a computer program, not only machine language code created by a compiler but also high-level language code that can be executed by a computer using an interpreter or the like may be included.

According to an embodiment, a method according to various embodiments of the present disclosure may be included in and provided by a computer program product (computer program product). The computer program product may be traded as a commodity between a seller and a buyer. The computer program product may be distributed in the form of a device readable storage medium (e.g., compact disc read only memory (CD-ROM)) or may be distributed (e.g., downloaded or uploaded) directly or online through an application Store (e.g., a Play Store TM) or between two user devices. For online distribution, at least a portion of the computer program product may be at least temporarily stored or created in a device-readable storage medium (e.g., memory, etc.) of a manufacturer server, an application store's server, or an intermediary server.

Unless explicitly stated or stated to the contrary with respect to the steps constituting the method according to the present invention, the steps may be performed in an appropriate order. The present invention is not necessarily limited to the order of description of the steps. All example or exemplary terms (e.g., etc.) used in the present invention are used only for detailed description of the invention, and the scope of the invention is not limited to the example or exemplary terms unless limited by the scope of the claims. In addition, it will be understood by those skilled in the art that various modifications, combinations, and variations can be made in accordance with design conditions and factors within the scope of the appended claims or equivalents thereof.

On the other hand, in the detailed description of the present invention, the specific embodiments have been described, but it will be apparent to those skilled in the art that various modifications can be made without departing from the scope of the invention.

Claims

1. A system for diagnosing dementia using story-based question-answer assessment, comprising:

a data input unit (10) capable of inputting text data of an actual story;

a data memory (20) which stores character data inputted through the data input unit (10) and which is capable of storing dementia information of each subject;

A voice output unit (30) which can output voice data for diagnosis to the outside so that the subject can hear the data;

a voice receiver (40) for receiving a response uttered by the subject; and

a controller (50) connected to the data input unit (10), the data memory (20), the voice output unit (30) and the voice receiver (40), the controller (50) comprising a story data forming unit (51), an answer data forming unit (52) and a dementia judging unit (53), wherein the story data forming unit (51) automatically forms adapted story data, question data and correct answer data based on an actual story inputted in the form of text data, the answer data forming unit (52) is for receiving an answer of a subject and forming answer data, the dementia judging unit (53) confirms an error answer rate by comparing and analyzing the correct answer data and judges whether dementia and/or judges a dementia state based on the degree of the error answer rate, and the controller (50) controls the data memory (20), the voice output unit (30) and the voice receiver (40) so as to sequentially output the adapted story data and the question data and receive an answer of the question of the subject to the outputted question, and further store information of each of the detected subjects judged dementia,

The actual story is formed of a plurality of sentences as a story form centering on a person whose subject is known and an event related to the person,

the adaptation story data is data formed by:

extracting a part of the words in the noun form included in the actual story and changing the extracted words in the noun form to words in other noun forms of the same or similar type to form an adapted story, and then converting the adapted story into voice data,

the problem data is data formed by:

taking words in the form of altered nouns included in the adaptation story as correct answers, and forming a plurality of questions related to the adaptation story, and converting the questions into voice data,

the correct answer data is data formed by:

extracting a correct answer corresponding to the question, and converting the extracted correct answer into voice data,

the answer data is voice data formed by:

answers to the subject to the questions are received and stored by the speech receiver (40).

2. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

The question consists of 3 to 12 questions, which are questions related to whether or not the event, behavior, scenario, and content shown in the adaptation story are true, and which are a form in which words in the form of nouns of the same or similar type as correct answers are included in wrong answers, one of a plurality of options is selected, and a question confirming the correct answer rate of the subject himself is configured as the last question.

3. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the actual story is a fairy tale story or a alley story known to the subject.

4. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the adaptation event consists of 5 to 15 sentences and comprises words of 20 to 50 different noun forms and comprises at least one additional sentence made by man, which is different from the content of the actual story and is related and connected to the content of the adaptation story and is placed at the end of the adaptation story.

5. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

The controller (50) receives a story notification text notifying an output of the adapted story for prompting a presentation of a problem related to the adapted story after the output of the adapted story, and converts the story notification text into voice data, and the control section (50) controls the voice output section (30) to output the story notification text in advance of outputting the adapted event.

6. The system for diagnosing dementia using story-based question-answer assessment of claim 5,

the adaptation story is a fairy tale story or a alley story for which the subject is known,

the story notification text places a sentence at the beginning that includes the title of the fairy tale story or the alley story presented as the adapted story.

7. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the controller (50) receives a question notification text for notifying that a predetermined number of questions are to be posed, converts the question notification text into voice data, and controls the voice output (30) to output the question notification text after the output of the adapted story is completed.

8. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the dementia determination unit (53) is configured to:

if the wrong answer rate is higher than at least 25%, dementia is judged, if the wrong answer rate is 25% to 40%, mild dementia is judged, and if the wrong answer rate is higher than at least 40%, severe dementia is judged.

9. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

preparing and respectively inputting a plurality of said actual stories and forming a plurality of adapted stories, and forming questions and correct answers associated with each of said adapted stories,

the controller (50) further includes:

a known story confirmation unit (54) that outputs a story confirmation text by voice to a subject to confirm which of a plurality of adapted stories are known, before outputting the adapted event; and

a story selection section (55) that selects an adapted story that is confirmed to be known to the subject to be examined, so as to output the adapted story confirmed by the known story confirmation section (54).

10. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the controller (50) further includes:

using a story confirmation unit (56) that, prior to outputting the adaptation event, outputs speech to the subject to confirm which of a plurality of adaptation stories was previously used for diagnostic examination; and

and a story excluding section (57) that excludes from the output the adapted story confirmed as previously used by the used story confirming section (56).

11. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the system for diagnosing dementia further comprises:

a display (60) connected to the controller (50) and capable of displaying text data on a screen; and

a text input device (70) connected to the controller (50) and capable of modifying text data by inputting text,

the controller (50) further includes:

a story confirmation and modification part (58) for displaying the adapted story on the display (60) in the form of text data so that a producer or an operator can confirm and manually modify the adapted story through the text input device (70).

12. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the system for diagnosing dementia further comprises:

the controller (50) further includes:

and a question confirmation and modification unit (59) for displaying the question on the display (60) in the form of text data so that a producer or an operator can confirm the question and manually modify the question by the text input unit (70).

13. The system for diagnosing dementia using story-based question-answer assessment of claim 1,

the system for diagnosing dementia further comprises:

a data transmitter (80) connected to the data memory (20) and controlled by the controller (50), and transmitting dementia information of each subject stored in the data memory (20) to an external device.

14. A method for diagnosing dementia using story-based question-answer assessment, comprising:

An actual story preparation step (S1) of preparing an actual story formed in a story form, composed of a plurality of sentences, centering on a person whose subject is known and an event related to the person;

an actual story input step (S2) of converting the prepared actual story into text data and inputting the text data to a diagnostic device;

an adaptation story forming step (S3) of extracting a part of words in noun form included in the input actual story and changing the extracted words in noun form to words in other noun form of the same or similar type to form an adapted adaptation story;

a question forming step (S4) of taking words in the form of altered nouns included in the adaptation story as correct answers and forming a plurality of questions in the form of text data related to the adaptation story;

a voice data conversion step (S5) of converting the story-adapted text data and the question text data into voice data;

a story-adapting output step (S6) of outputting voice data of the story-adapting to the outside to be provided to a subject;

a question outputting step (S7) of outputting the voice data of the question to the outside in order to raise the question to the subject after completing the output of the recomposition to the subject;

An answer receiving step (S8) of receiving an answer to each of the questions from the subject; and

and a dementia judging step (S9) of confirming a wrong answer rate by comparing and analyzing the received answer and the right answer, and judging whether dementia is caused and/or judging the dementia state according to the degree of the wrong answer rate.

15. A computer-readable recording medium, characterized in that a program for executing the method of claim 14 on a computer is recorded.