KR101804389B1 - System and method for autonomically testing dementia - Google Patents

System and method for autonomically testing dementia Download PDF

Info

Publication number
KR101804389B1
KR101804389B1 KR1020150163885A KR20150163885A KR101804389B1 KR 101804389 B1 KR101804389 B1 KR 101804389B1 KR 1020150163885 A KR1020150163885 A KR 1020150163885A KR 20150163885 A KR20150163885 A KR 20150163885A KR 101804389 B1 KR101804389 B1 KR 101804389B1
Authority
KR
South Korea
Prior art keywords
language
information
inquiry
basic
unit
Prior art date
Application number
KR1020150163885A
Other languages
Korean (ko)
Other versions
KR20170059664A (en
Inventor
곽용진
최지선
박재은
이지연
Original Assignee
주식회사 이르테크
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 주식회사 이르테크 filed Critical 주식회사 이르테크
Priority to KR1020150163885A priority Critical patent/KR101804389B1/en
Publication of KR20170059664A publication Critical patent/KR20170059664A/en
Application granted granted Critical
Publication of KR101804389B1 publication Critical patent/KR101804389B1/en

Links

Images

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/40Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4076Diagnosing or monitoring particular conditions of the nervous system
    • A61B5/4088Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia
    • G06F17/2705
    • G06F17/271
    • G06F17/2755
    • G06F17/2785
    • G06F19/34
    • G06F19/363
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/26Speech to text systems

Abstract

The present invention relates to an interactive inquiry unit for generating basic inquiry information of a subject to be inspected and for deriving basic information for the basic inquiry corresponding to the basic inquiry information; A language analyzing unit for analyzing the basic speech information and measuring a language disorder corresponding to each language ability item, and comparing the corpus information and the diagnostic evaluation information with the language disorder to determine the result of the inquiry including the disorder, Discloses a language obstacle autonomous inspection system and method for early diagnosis of a language disorder that induces utterance of a testee for each scenario by referring to a type of obstacle if the judgment is judged to be pending .

Description

[0001] SYSTEM AND METHOD FOR AUTOMATICALLY TESTING DEMENTIA [

More particularly, the present invention relates to a technique for discriminating the presence or absence of a language disorder, such as primary progressive aphasia and mild cognitive impairment accompanied by dementia.

A conventional neurological approach based diagnosis of dementia was performed through blood tests and brain imaging tests using MRI (Magnetic Resonance Imaging) or Positron Emission Tomography (PET).

The neuropsychological approach or the neurolinguistic approach based dementia diagnosis was performed through a neuropsychological test which indirectly grasps the cognitive function of the brain.

Among the aforementioned tests, the determinants for diagnosing dementia are physical examinations, blood tests and brain imaging tests.

Typically, dementia and accompanying language disorders are classified into Primary Progressive Aphasia (PPA) and Mild Cognitive Impairment (MCI), which can be diagnosed through the results of brain imaging or pathological features, In order to make an accurate diagnosis, it was necessary to identify the characteristics of speech disorders first and make an early diagnosis.

However, conventional diagnosis of dementia has been problematic in that it is diagnosed by brain image examination, analysis of brain waves by stimulation and analysis of the questionnaire by questionnaires, and posterior diagnosis is performed in which examination is performed.

In addition, since the conventional diagnosis of dementia is dependent on expensive equipment such as a brain imaging test or is an indirect method based on cognitive ability evaluation, it is difficult for the user to recognize the dementia due to a burden of cost or an environmental burden, There is a problem that accessibility for diagnosis is not easy and the user is difficult to inspect for himself.

1. Korean Registered Patent No. 10-0680232 (2007.02.01.)

The present invention overcomes a physical inspection environment, autonomously provides a testee's self-diagnosis, analyzes the basic speech information corresponding to basic query information to provide a language barrier, And methods.

The present invention provides a system and method for verbal language disability autonomous inspection for various types of disabilities, which measures language impairment based on language analysis based on language phenomena classification and language ability items.

The present invention provides an early diagnosis of dementia to a testee using a corpus information and diagnostic evaluation information, and provides a dynamic documentary scenario to a testee in case of suspected dementia or language impairment judgment, inducing a personalized speech, Provided is an autonomous language inspection system and method for providing an interactive test, not an inspection, to reduce the error of language failure judgment.

An autonomous language disability autonomous inspection system for early diagnosis of a language disorder according to the present invention includes: an interactive paper-making unit for generating basic question information of an examinee and deriving basic information about the basic question information; A language analyzing unit for analyzing the basic speech information and measuring a language disorder corresponding to each language ability item, and comparing the corpus information and the diagnostic evaluation information with the language disorder to determine the result of the inquiry including the disorder, And the interactive questionnaire section refers to the suspected disorder type if the judgment is judged to be suspended, and induces the testee's utterance for each scenario.

The language barrier autonomous examination system may further include a document terminal unit for providing basic query information to the testee and receiving the basic language information.

The interactive inquiry unit searches the presence or absence of the test history information of the examinee, generates inquiry inquiry information based on the inspection history information if the inspection history information exists, and generates basic inquiry information if the inspection history information does not exist .

The interactive paper inspection section can generate the paper query information for each scenario having the lexical diversity, time complexity, expressiveness, and topic continuity.

The language analyzing unit may include a language processing module for converting the basic speech information into text and processing the text, analyzing the text based basic speech information in a grammatical unit, and processing the parsed basic speech information; A language phenomenon classification module for classifying the basic speech information processed in the grammatical unit into a language phenomenon; and a language phenomenon classification module for classifying the basic speech information classified into the language phenomenon and the language disorder corresponding to the language ability item on the basis of the evaluation value referred to the language phenomenon classification And a language ability determination module for measuring the degree of the language.

The corpus information may include statistical processed speech processing information by filtering the language impairment analyzed by the language analyzing unit and processing the statistically processed language impairment candidate information or the result of the inquiry.

A method for operating a language barrier autonomous inspection system for early diagnosis of a language disorder according to the present invention includes the steps of searching an inspection history information of a testee in an interactive paper inspection unit, Generating basic query information if the test history information of the testee does not exist; Analyzing the basic speech information responded to by the testee in response to the test history information or the basic query information in the language analyzing unit and measuring a language disorder corresponding to each language ability item; Comparing the corpus information and the diagnostic evaluation information with the language impairment level in the corpus judgment section to judge the result of the inquiry such as the presence or absence of the disorder, the type of the disorder and the judgment pending, and, , Selecting the query type for each suspicious factor, and searching the scenario for each query type to generate the inquiry query information.

The present invention enables automatic diagnosis by the testee through the self terminal of the language barrier autonomous inspection system or the medical terminal such as the desktop and the smart phone, and enables the early diagnosis of dementia having no intervention of any other person, .

According to the present invention, it is possible to analyze an utterance classified by a wide range of disabilities necessary for diagnosis of a suspected language disorder from a testee through a language analyzer in real time, thereby measuring or collecting accurate and accurate diagnostic data.

The present invention can provide a diagnosis of dementia to a testee using corpus information and diagnosis evaluation information or diagnosis record information of a testee so that it is possible to identify and diagnose a wide range of language disorder types and to update diagnosis record information of a testee can do.

The present invention can provide a dynamic documentary scenario to a testee through an interactive paper-testing unit in the case of suspected dementia or language impairment, thereby providing an interactive test rather than a one-way one-way test, .

FIG. 1 is a block diagram showing a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder of the present invention.
2 is a detailed block diagram of the autonomous language testing system of FIG. 1;
FIG. 3 is a flowchart showing an example of generating the corpus information and the process of the subject's utterance according to the present invention.
4 is a flowchart illustrating an operation method of a conventional language barrier autonomous inspection system.
5 is a flowchart illustrating an operation method of a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder according to a first embodiment of the present invention.
6 is a flowchart illustrating an operation method of a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder according to a second embodiment of the present invention.
7 is a flowchart illustrating an operation method of a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder according to a third embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and accompanying drawings, but the present invention is not limited to or limited by the embodiments.

FIG. 1 is a block diagram illustrating a language barrier autonomous testing system for early diagnosis of a dementia disease accompanied by a language disorder according to the present invention. The language barrier autonomous testing system 100 includes an interactive questionnaire 110, (120) and a paper judging section (130).

The language barrier autonomous inspection system 100 generates basic query information through the interactive questionnaire 110, derives basic speech information corresponding to the basic query information, analyzes the basic speech information through the language analyzer, The language barrier corresponding to each capability item is measured, and the coroner judgment unit 130 compares the corpus information and the diagnostic evaluation information with the language impairment, and judges the result of the inquiry including the obstacle, the type of the obstacle, and the judgment pending.

If it is determined that the judgment is pending, the interactive questionnaire section 110 generates inquiry query information for inducing utterance of the examinee for each scenario by referring to the failure type. Hereinafter, with reference to FIG. 2, the interactive paper inspection unit 110, the language analysis unit 120, and the paperweight determination unit 130 will be described in detail.

FIG. 2 is a detailed block diagram of the language barrier autonomous inspection system of FIG. 1, in which the interactive questionnaire 110 generates basic query information of a testee and derives basic speech information corresponding to basic query information.

The basic query information includes test item data for checking whether a dementia disease accompanies speech disorder, test item data for discriminating the type of language disorder, and test item data for understanding the utterance intention.

The basic utterance information may include data of recorded utterances or voices of the examinee in response to the basic query information.

The inquiry query information includes more detailed item data than the basic query information, and includes test item data for determining whether the testee has a language disorder. The questionnaire utterance information includes data on the voice of the examinee responding to the inquiry query information .

The inquiry query information may include test item data for analyzing the basic speech information and inducing speech of the examinee for each scenario, and the following description will be referred to.

The language barrier autonomous inspection system 100 may further include a document terminal 300 that provides basic query information of a testee and receives basic speech information.

The document terminal unit 300 may be the mobile communication terminal 310 or the desktop 320 or may be a terminal included in the language barrier autonomous inspection system 100. The mobile communication terminal 310 or the desktop 320 may be a testee's terminal.

The mobile communication terminal 310 may be a portable terminal, a digital broadcasting terminal, a personal digital assistant (PDA), an ATM, a smart phone, an International Mobile Telecommunication 2000 (IMT-2000) terminal, a Wideband Code Division It will be obvious that the present invention can be applied to all information communication devices and multimedia devices such as a multiple access terminal, a UMTS (Universal Mobile Telecommunication Service) terminal, a smart watch, and the like.

The document terminal unit 300 can receive basic query information from the interactive paper-testing unit through the wired / wireless network 200, and can receive the basic-speech information based on the input of the testee.

The documentant terminal unit 300 provides basic query information to the examinee and interlocks with the terminal of the person to be tested installed with the application or application program in order to receive the basic speech information in response to the basic personality information in response to the basic query information, Can be constructed.

The medical consultation terminal unit 300 can directly diagnose the dementia because it can be self-diagnosed directly by the testee, has no intervention of other persons, has no irritation, and is easy to access.

The documentant terminal unit 300 can display a login information request via a display to confirm the identity of the examinee and can display a personal information request for creating a new account and the examinee can request a login information or a personal information request Corresponding login information or personal information can be input through the document terminal unit 300. [

The interviewer terminal unit 300 may preliminarily process the basic speech information and the personal information of the subject to be transmitted to the interactive paperweighting unit. The interactive papering unit 110 searches for presence or absence of the examination history information of the subject, And if there is no test history information, the basic query information can be transmitted to the document terminal unit 300. [0053] FIG.

The language analyzer 120 analyzes the basic speech information and measures the language barrier corresponding to each language ability item.

The language analyzing unit 120 is for analyzing the meaning of utterances as well as the natural language processing for the basic utterance information. The language analyzing unit 120 includes a language processing module 121, a language phenomenon classification module 122, a language ability determination module 123 The language analysis of the basic speech information can be performed through the subdivided modules.

The language processing module 121 may convert the speech-based basic speech information into text, process the text, and analyze the text-based basic speech information in grammatical units.

The language processing module 121 can analyze and process text based basic speech information in grammatical units such as morphological analysis, syntax analysis, semantic analysis, discourse analysis, and phonetic analysis. For example, in the case of a discourse analysis, if the text-based basic utterance information is "Young-hee was pretty, but he was not habitual ", the language processing module 121 may correct the " Missing restoration and misrecognition can be analyzed and analyzed, and 'Yeonhee' and 'He' can be analyzed in relation to notifications, and the possibility of errors in the notified relationship can be analyzed.

The language phenomenon classification module 122 can classify the basic speech information processed in a grammatical unit into a language phenomenon.

The language phenomenon classification module 122 classifies the basic speech information processed in a grammatical unit into a language such as vocabulary diversity, usage rate of each part of speech, functional language utilization rate, subordinate rate per sentence, connection ratio per sentence, syntax tree depth, syntax complexity, It can be classified as a phenomenon. For example, when classifying the linguistic phenomena for lexical diversity, the linguistic phenomenon classification module 122 compares the number of uttered vocabulary classes and the number of uttered vocabularies to classify which kind of vocabulary is included in the testee's utterance Can be calculated and classified.

Table 1 is a table showing the analysis items of morphological / language impairments that can be used for the classification of language phenomena.

Serial Number Qualitative characteristics Meaningful qualities
(In English, based on p value)
Qualifications (original language) Qualifications (Korean) Applicable in Korean concept
One Vocabulary Usage Pattern Words word 2 Vocabulary Usage Pattern 2 Nouns noun # Nouns / # words 3 Vocabulary Usage Pattern Verbs verb # Verbs / # words 4 Vocabulary Usage Pattern One Noun-verb ratio The ratio of nouns to verbs # Nouns / # verbs 5 Vocabulary Usage Pattern One Noun ratio The ratio of nouns to the sum of nouns and verbs # Nouns / (# nouns + # verbs) 6 Vocabulary Usage Pattern Inflected verbs Reflexive verb X: Since Korean is a pseudonym, there is no distinction between refraction type and non-refraction type in the verb. 7 Vocabulary Usage Pattern Light verbs Functional verb ○: Korean function verb list required # Light verbs / # verbs 8 Vocabulary Usage Pattern Determiners Qualifier △: Since there are no qualifiers such as articles in Korean, # Determiners / # words 9 Vocabulary Usage Pattern 2 Demonstratives Geisha # Demonstratives / # words 10 Vocabulary Usage Pattern Prepositions preposition X: No preposition in Korean # Prepositions / # words 11 Vocabulary Usage Pattern Adjectives adjective # Adjectives / # words 12 Vocabulary Usage Pattern 2 Adverbs adverb # Adverbs / # words 13 Vocabulary Usage Pattern One Pronoun ratio Pronoun rate # Pronouns / (# nouns + # pronouns) 14 Vocabulary Usage Pattern Function words Function word ○: A list of functional word categories in Korean is required # Function words / # words 15 Vocabulary related variables 2 Frequency Average frequency Mean frequency of all words appearing in the frequency norms 16 Vocabulary related variables 2 Noun frequency Noun average frequency Mean frequency of nouns appearing in the frequency norms 17 Vocabulary related variables One Verb frequency Average frequency of verbs Mean frequency of verbs appearing in the frequency norms 18 Vocabulary related variables Imageability Imaginary castle Mean imageability of all words appearing in the imageability norms 19 Vocabulary related variables Noun imageability Noun imagination castle Mean imageability of nouns appearing in the imageability norms 20 Vocabulary related variables One Verb imageability Verb image castle Mean imageability of verbs appearing in the imageability norms 21 Vocabulary related variables Age of acquisition Age of acquisition Mean age of acquisition norms 22 Vocabulary related variables Noun age of acquisition Noun acquisition age Mean age of acquisition nouns 23 Vocabulary related variables Verb age of acquisition The age of learning the verb Mean age of acquisition norms 24 Vocabulary related variables 2 Familiarity Familiarity Mean familiarity of all words 25 Vocabulary related variables 2 Noun familiarity Noun familiarity Mean familiarity of nouns 26 Vocabulary related variables Verb familiarity Verb familiarity Mean familiarity of verbs appearing in the familiarity norms 27 Fluency Type-Token Ratio (TTR) Vocabulary diversity # Unique word types / # words 28 Fluency 2 Word length Word length Mean number of letters in each words 29 Fluency Fillers Kangwon # Fillers / # words 30 Fluency Um (whatever) High frequency kansas # Occurrences of / # words 31 Fluency Uh (er) High frequency kansas # Occurrences of / # words 32 Fluency 2 Speech rate Ignition speed # Words uttered / total time in minutes 33 Vocabulary Usage Pattern Investigation ++: Reflects the Korean part-time system 34 Vocabulary Usage Pattern Research ++: Reflects the Korean part-time system 35 Vocabulary Usage Pattern A good mother ++: Major grammar stemming in Korean 36 Vocabulary Usage Pattern Mother-in-law ++: Major grammar stemming in Korean

(Legend: ○ possible, △ fix applicable, × impossible, ++ additional required)

Table 2 is a table showing the syntactic language disorder analysis items that can be used for the classification of language phenomena.

Serial Number endowment
characteristic
Meaningful qualities
(English,
p value basis)
Qualifications (original language) Qualifications (Korean) Applicable in Korean concept
One Syntactic
complexity
Sentences sentence
2 Syntactic
complexity
T-units T unit (minimal terminal unit) A clause and all of its dependent clauses
3 Syntactic
complexity
One Clauses section A structure consisting of at least a subject and a finite verb
4 Syntactic
complexity
Coordinate phrases Equatorial Connection A phrase immediately before a coordinating conjunction
5 Syntactic
complexity
Complex nominals Compound noun A noun phrase, clause, or gerund that stands in a noun
6 Syntactic complexity Complex T-units Complex T unit A T-unit which contains a dependent clause 7 Syntactic complexity Verb phrases Verbal A phrase consisting of at least a verb and its dependents 8 Syntactic complexity Dependent clauses subordinate A clause which could not form a sentence on its own 9 Syntactic complexity One Mean length of sentence Average length of sentence 10 Syntactic complexity One Mean length of clause Average length of section 11 Syntactic complexity Mean length of T-unit T unit
Average length
12 Syntactic complexity Dependent clauses per clause Ratio of subordinate clauses to clauses 13 Syntactic complexity Dependent clauses per T-unit Percent of subordinate clauses for T units 14 Syntactic complexity Verb phrases per T-unit Phrase ratio for T units 15 Syntactic complexity Clauses per sentence For the sentence
Percentage of clause
16 Syntactic complexity Clauses per T-unit Percentage of clauses for T units 17 Syntactic complexity Complex T-units per T-unit The ratio of complex T units to T units 18 Syntactic complexity Coordinate phrases per T-unit Ratio of peer connections to T units 19 Syntactic complexity Complex nominals per T-unit Percentage of compound noun phrases for T units 20 Syntactic complexity One T-units per sentence Percentage of T units for sentences 21 Syntactic complexity Coordinate phrases per clause Percentage of peer connection to section 22 Syntactic complexity Complex nominals per clause Ratio of compound nouns to clauses 23 Syntactic complexity Tree height Depth of parsing tree Height of the parse tree 24 Syntactic complexity One Total depth Yngve (1960) Method of parsing The depth of the diagram Total Yngve depth 25 Syntactic complexity Max depth Yngve (1960) Method of parsing The maximum depth of a tree Maximum Yngve depth 26 Syntactic complexity One Mean depth Yngve's parsing method Depth average of tree Mean Yngve depth 27 Syntactic complexity Idiomatic phrases per T-unit Ratio of idioms to T units ++

(Legend: ○ possible, △ fix applicable, × impossible, ++ additional required)

Referring to Table 1 and Table 2, the language phenomenon classification module 122 can classify the basic speech information processed in a grammatical unit into the types, meaning, and syntactic language disorder analysis items used in the language phenomenon classification.

The language ability determination module 123 can measure the language disorder corresponding to each language ability item based on the basic speech information classified as the language phenomenon and the evaluation value referred to the language phenomenon classification.

The language ability determination module 123 analyzes language difficulties corresponding to language ability items such as language fluency, vocabulary usage pattern, semantic relation, sentence structure, and tense processing ability based on basic speech information and evaluation values classified as language phenomena Can be measured.

The evaluation value may include a numerical value of the diagnostic evaluation information as shown in Table 3 shown below, and the machine such as xml and rdb may include the diagnostic evaluation reference information based on the toxicity type as shown in Table 4 below, It can be used to classify by ability item, obstacle type, type of obstacle and judgment judgment.

Table 3 is a table exemplarily showing the diagnostic evaluation information.

Patient group Diagnostic criteria Exclusion criteria Primary
Progression
aphasia
■ The most prominent clinical features are language disorders
■ Language problems cause obstacles to daily living.
■ The most prominent damage at the onset and in the early stages of the disease is aphasia
■ Damage pattern due to non-degenerative nervous system problems or disease
Cognitive impairment due to psychiatric problems
■ Anecdotal memory, visual memory, visual perception problems
■ Behavior problems that are prominent in the early stages of illness
Mild cognitive impairment Memory ■ Subjective cognitive impairment complaint of patient or guardian
■ Overall normal cognitive function
■ Maintain the ability to perform everyday activities
■ Neuropsychological assessment of objective memory impairment compared with age
■ Does not meet diagnostic criteria for dementia
■ If you are diagnosed with dementia
Neurological problems that lead to cognitive dysfunction (stroke, brain tumor, traumatic brain injury, epilepsy) and psychiatric problems
■ Geriatric Depression ■ Scale (GDS) -15 scores greater than 10 points for depression
■ Systemic illness with brain damage (metabolic disorders, endocrine disorders, toxic problems, infectious diseases)
■ History of alcoholism and / or drug addiction
■ exhibits abnormal performance in Mini-Mental State Examination (MMSE); If the number of years of education is less than 11 years, it is 22 or less. If the number of years of education is more than 11 years, it is considered to be abnormal.
Non-memory ■ Dysfunction in other areas (eg, executive function, language function) other than memory
■ Everyday function is maintained
■ Can not be classified as dementia
■ Except for objective memory impairment, same as memory impairment cognitive impairment cognitive impairment criteria

Table 4 is an exemplary table showing the diagnostic evaluation reference information of a machine such as xml and rdb.

Figure 112015113901926-pat00001

The paperweight determining unit 130 compares the corpus information and the diagnostic evaluation information with the language impairment to determine the result of the paperwork such as the presence of the impairment, the type of the impairment, and the judgment pending.

The paperweight determining unit 130 may compare the language impairment corresponding to the language ability item with the corpus information and the diagnostic evaluation information to determine the result of the paperwork such as the presence or absence of the disorder, the type of the disorder, and the judgment pending.

The corpus information may include statistically processed document processing information by filtering the language impairment analyzed by the language analyzing unit and processing the statistically processed language impairment candidate information or the survey result.

FIG. 3 is a flow chart showing an example of generating the corpus information and the process of the subject's utterance according to the present invention. The interactive paper detector 110 receives basic subject information of the subject from the document terminal 300, Unit 120 to transmit the basic speech information.

The language analysis unit 120 analyzes the basic speech information in a grammatical unit such as morphological analysis, syntax analysis, semantic analysis, discourse analysis, and phonetic analysis. The language analysis unit 120 analyzes vocabulary diversity, usage rate of each part, utilization rate of function word, Language disorder, such as rate, syntax tree depth, syntax complexity, and idiomatic / idiomatic usage rates.

The language analyzer 120 may generate and store the corpus information including the statistically processed document processing information by filtering the language impairment and processing the statistically processed language impairment candidate information or the survey result.

The candidate for language impairment information is information classified into an item of an error filtering for a morphological level, a vocabulary, a syntax structure, a sentence relation, and a speech.

Referring back to FIG. 2, the paperweight determining unit 130 may determine the paperback result by integrally comparing the corpus information, the diagnostic record information, and the language impairment. The diagnostic record information may be the information on which the past diagnosis of the examinee is recorded, and may be included in the diagnostic evaluation information.

The paperweight determining unit 130 can determine the paperwork result by integrally comparing the corpus information, the pathology information, and the language impairment. The pathological information may be information formulated by expert opinion information related to diagnosis of dementia or experimental results such as clinical experiments, and may be included in diagnostic evaluation information.

If it is determined that the judgment is pending, the interactive questionnaire section 110 generates inquiry query information for inducing utterance of the examinee for each scenario by referring to the temporary obstacle type. Temporary disability types are not judged to be language disorders, but may include suspected language disorders.

The interactive inquiry unit 110 can generate inquiry query information for each scenario having the purpose of utterance induction such as vocabulary diversity, time complexity, expressiveness, and topic continuity.

The interactive paper inquiry unit 110 can generate inquiry query information for inducing utterance through an event query, an order query, a detailed query, and a sequence manipulation query for time complexity. For example, the interactive paperwork unit 110 may be able to say, 'What are the three most recent memorabilia, what is the oldest thing? What does the first thing have to do with the second? ' Can generate query query information that induces utterance for time complexity as shown in Fig.

The interactive inquiry unit 110 can generate inquiry query information for each scenario according to the evaluation items for the type of the language disorder for inducing utterances, and Table 5 is a table showing evaluation items for each type of the language disorder.

object

language
Stratification task
Primary progressive aphasia (PPA)  Mild cognitive impairment (MCI) Normal old age
Analogy Meaningful Ignition
Tribal
Vegan Dry
Form - Semantic noun
name
Waiting
■ General and various name demonstration ■ Degradation of face-to-face name
■ Especially difficult to wait for a noun name
■ Degradation of name waiting capability
■ Speech Fluency (Category, Phoneme) ■ Upper category vocabulary replacement ■ Error generation that is semantically related to target words at face-to-face name waiting, inaccurate and slow response
■ Increased difficulty in controlled word association test

■ Obsessive-compulsive obsessive-compulsive only at ages 70 and above.
verb
name
Waiting
■ Especially difficult to wait for the name of the verb ■ Difficulties in proper verbs in context
Self-ignition ■ Meaning Adherence ■ difficulty in word calculation
■ Difficult to find words
■ Function error increases ■ Frequent rests in sentences
■ Long reaction time
■ Using incorrect vocabulary
■ Difficulty in generating proper nouns and low frequency words when calculating utterances
■ The frequency of use of verbs and pronouns is high when calculating utterances
■ When calculating utterances, difficulty in word calculation,
Syntactic word
Justice
doing
■ Descriptive power degraded ■ Defining Words
Self-ignition ■ Grammar simplification and error in language calculation
■ Room grammar
Slow horse speed
■ Grammar skills are generally conserved. Some studies observed mild grammatical disorders ■ Less ignition
■ difficult to start speaking
■ No grammar proof
■ Due to the difficulty of finding words
■ Speech Speed Declination
■ Reduced error self-correction capability
■ Grammar ability preservation
■ I prefer to calculate sentences of the right branch structure
■ Reduced syntactic diversity and complexity

■ The influence of aging on the versatility of syntactic structure (controversial)

The interactive questionnaire section 110 transmits the questionnaire query information constituted by scenarios to the document terminal unit 300 and receives the questionnaire speech information corresponding to the information of the door quality from the documentarian terminal unit 300. [

The language analyzing unit 120 and the inquiry determining unit 130 may be performed in the same manner as the method of analyzing and querying the inquiry query information for the basic query information described above, and a description thereof will be omitted.

Referring again to FIG. 1, the language barrier autonomous inspection system 100 includes a basic query base 141 in which basic query information is stored, a language base 143 in which information on the language capability item is stored, a corpus base 145 in which corpus information is stored, A diagnostic evaluation base 147 in which diagnostic evaluation information including pathological information or diagnostic record information of the examinee is stored, and a database 149 containing a scenario base in which document query information is stored for each scenario.

The interactive paper inquiry unit 110, the language analysis unit 120 and the paper inquiry unit 130 fetch the information stored in the database 140 to perform a query, a language analysis, and a result of the inquiry related to the utterance.

FIG. 4 is a flowchart illustrating an operation method of a conventional language barrier autonomous inspection system. FIG. 5 is a flowchart illustrating an operation of a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder according to the first embodiment of the present invention. As a flowchart showing the method, a comparison between FIG. 4 and FIG. 5 will be described.

Referring to FIG. 4, the conventional language disability autonomous inspection system can determine whether or not a language disorder occurs by referring to a knowledge base such as language analysis and ontology to analyze utterance intentions.

In the case of the conventional language obstacle autonomous inspection system, when the language impairment judgment is suspended, the query information can be generated using the language-analyzed information and the knowledge base.

Referring to FIG. 5, the autonomous language impairment inspection system of the present invention measures language impairment through language analysis based on language phenomena classification and language ability items, and judges whether language impairment is caused by referring to corpus information and diagnostic evaluation information can do.

The language barrier autonomous inspection system of the present invention can generate the inquiry query information for each scenario having the purpose of utterance induction such as vocabulary diversity, time complexity, expressiveness and topic continuity when the language obstacle judgment is suspended.

In contrast to the conventional language disability autonomous inspection system, the language disability autonomous inspection system of the present invention analyzes a language in a grammatical unit, analyzes language based on language phenomena classification and language ability items, measures language impairment, And can provide a dynamic documentary scenario to the testee, thereby reducing the error of the language trouble judgment.

6 is a flowchart illustrating an operation method of a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder according to a second embodiment of the present invention.

Referring to FIG. 6, the language barrier autonomous inspection system reads and generates basic query information of a testee from a basic query base, and derives basic speech information corresponding to the basic query information.

The language impairment self - examination system analyzes the basic speech information and measures the language impairment corresponding to each language ability item. Information about the language capability item can be read from the language base.

The autonomous language testing system compares the corpus information and the diagnostic evaluation information with the language impairment, and judges the result of the inquiry including the impairment, the type of the impairment and the judgment pending. The corpus information can be read out from the corpus base, the diagnostic evaluation information can be read out from the diagnostic evaluation base, and the diagnostic evaluation information can include pathological information or diagnostic record information of the examinee.

In the case of the language obstacle autonomous inspection system, if it is determined that the judgment is suspended, the system generates the inquiry query information for inducing the testee's utterance for each scenario by referring to the type of the obstacle.

The language barrier autonomous inspection system searches for presence or absence of test history information of the testee, generates inquiry query information based on the test history information if the test history information exists, and generates basic query information if the test history information does not exist can do.

7 is a flowchart illustrating an operation method of a language barrier autonomous inspection system for early diagnosis of a dementia disease accompanied by a language disorder according to a third embodiment of the present invention.

Referring to FIG. 7, the autonomous language disability inspection system can request login information to confirm the identity of a testee, request personal information for account creation, The login information or the personal information can be received.

The language barrier autonomous inspection system can search the test history information of the testee by referring to the personal information of the testee. The test history information can be stored in the diagnostic evaluation base.

The language barrier autonomous inspection system generates inquiry query information based on the inspection history information if the inspection history information exists, and may generate the basic query information if the inspection history information does not exist.

The basic query information can be stored in the basic query base, and the query query information can be generated through analysis of the basic speech information corresponding to the basic query information.

The language disability autonomous inspection system analyzes the test history information or the basic language information to measure the language disorder corresponding to each language ability item, compares the corpus information and the diagnostic evaluation information with the language disorder, The results of the hold, etc., are determined.

Information on language capability items can be stored in the language base, corpus information can be stored in the corpus base, and diagnostic evaluation information can be stored in the diagnostic evaluation base.

In the case of the language obstacle autonomous inspection system, if it is determined that the judgment is pending, the inquiry query information for inducing the testee's utterance for each scenario can be generated with reference to the obstacle type. If the normal obstacle or the obstacle is determined, the result of the inquiry is notified to the examinee , And the results of the interviews can be stored in the diagnostic evaluation base.

In the case of the language obstacle autonomous inspection system, if it is judged to be a judgment pending, it is possible to sort by the language suspicion factor, select the question type for each suspicious factor, and generate the inquiry query information by searching the scenario for each query type.

The language barrier autonomous inspection system collects the constructed inquiry query, generates inquiry query information, provides it to the examinee, and receives the query speech information corresponding to the inquiry query information.

The analysis of the speech utterance information and the determination of the re-examination result will be made by referring to the above-described analysis of the basic utterance information and the method of judging the result of the inquiry.

100: Language Disorder Autonomous Inspection System
110: Interactive paperweight
120: Language Analysis Department
130:
140: Database
200: Wired and wireless network
300:

Claims (7)

A system for automatically diagnosing language impairment for early diagnosis of a language disorder,
An interactive type inquiry unit for generating basic inquiry information of the testee in an interactive form and for deriving basic information for the basic inquiry information;
A language analysis unit for classifying the basic speech information as a language phenomenon and measuring a language disorder corresponding to each language ability item;
A corpus determination unit for comparing corpus information, diagnosis evaluation information, and language impairment to judge the result of the inquiry such as a fault, a fault type,
The interactive inquiry unit generates inquiry inquiry information for referring to the suspicious obstruction type and inducing utterance of the examinee for each scenario if the judgment is judged as suspending the judgment,
Characterized in that the test subject is provided with a dynamic documentary scenario to induce utterance, and an interactive self-diagnostic test is provided.
The method according to claim 1,
And a speech terminal unit for providing the basic question information to the testee and inputting the basic speech information.
The method according to claim 1,
The interactive inquiry unit searches for the presence or absence of the test history information of the examinee, generates inquiry inquiry information based on the inspection history information if the inspection history information exists, and generates basic inquiry information if the inspection history information does not exist Autonomous language screening system.
The method according to claim 1,
The interactive paper inspection unit is a language disability autonomous inspection system that generates document query information for each scenario having the purpose of inducing vocabulary diversity, time complexity,
The method according to claim 1,
The language analyzing unit,
A language processing module for converting and processing the basic speech information into text, analyzing the text based basic speech information in a grammatical unit, and processing the parsed basic speech information;
A language phenomenon classification module for classifying the basic speech information processed in the grammatical unit into a language phenomenon;
And a language ability determination module that measures a language impairment corresponding to each language capability item based on the basic speech information classified into the language phenomenon and the evaluation value referred to the language phenomenon classification.
The method according to claim 1,
Wherein the corpus information includes a statistically processed speech processing information by filtering the language impairment analyzed by the language analyzing unit and processing the statistically processed language impairment candidate information or the result of the inquiry.
A method of operating a language barrier autonomous inspection system for early diagnosis of a language disorder,
If the test history information of the examinee is present, the test history information is transmitted to the language analyzing unit. If the test history information of the testee does not exist, ≪ / RTI >
Classifying the basic speech information, which is answered by the testee in response to the test history information or the basic query information, into a language phenomenon in the language analyzing unit and measuring a language disorder corresponding to each language ability item;
Comparing the corpus information, the diagnostic evaluation information, and the language impairment by the corporeal judgment unit to determine the result of the inquiry including the impairment, the type of the impairment and the judgment pending, and
Selecting a question type for each suspicious factor, and searching for scenarios for each question type to generate inquiry query information, if the inquiry hold is determined in the interactive questionnaire part,
A method for operating a language barrier autonomous inspection system, characterized by providing a dynamic documentary scenario to a subject to induce utterance, and to provide an interactive self-diagnostic test.
KR1020150163885A 2015-11-23 2015-11-23 System and method for autonomically testing dementia KR101804389B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020150163885A KR101804389B1 (en) 2015-11-23 2015-11-23 System and method for autonomically testing dementia

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020150163885A KR101804389B1 (en) 2015-11-23 2015-11-23 System and method for autonomically testing dementia

Publications (2)

Publication Number Publication Date
KR20170059664A KR20170059664A (en) 2017-05-31
KR101804389B1 true KR101804389B1 (en) 2017-12-04

Family

ID=59052477

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020150163885A KR101804389B1 (en) 2015-11-23 2015-11-23 System and method for autonomically testing dementia

Country Status (1)

Country Link
KR (1) KR101804389B1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102064682B1 (en) 2018-09-21 2020-01-09 이화여자대학교 산학협력단 Method for fast mapping ability measurement using a quick incidental learning, recording medium and apparatus for performing the method
KR20200085041A (en) 2019-01-04 2020-07-14 순천향대학교 산학협력단 Language rehabilitation based vocal voice evaluation apparatus and method thereof
KR20210102620A (en) * 2020-02-12 2021-08-20 이화여자대학교 산학협력단 Video-based language disorder analysis system, method and program for performing the analysis

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102041848B1 (en) * 2017-12-29 2019-11-08 동국대학교 산학협력단 Early detection system of depression, anxiety, premature dementia or suicide by Artificial intelligence-based speech analysis
KR101894722B1 (en) * 2018-03-12 2018-10-04 미디어젠(주) Health anomaly prediction system using speech disorder diagnosis and Method
KR102142957B1 (en) 2018-08-24 2020-08-12 주식회사 리스트 Machine-lead conversation service method and system
KR102254685B1 (en) * 2018-12-28 2021-05-21 주식회사 보듬 System for screening early mild cognitive impairment
KR102314572B1 (en) * 2019-06-17 2021-10-18 한림대학교 산학협력단 Method and system for assessing language ability based on artificial intelligence
KR102245941B1 (en) * 2019-09-17 2021-04-29 보리 주식회사 Continuous conversation-based developmental disability testing system and method
KR102297111B1 (en) * 2019-12-31 2021-09-03 광주과학기술원 Anomalies Classification System and Method Based on Deep Neural Network Algorithm
KR102533368B1 (en) * 2020-05-29 2023-05-17 보리 주식회사 Method and system for analyzing fluency using big data
KR102237539B1 (en) * 2020-07-31 2021-04-07 이승연 System and method for determining demendia and congnitive ability using voice conversation analysis
KR102444012B1 (en) * 2020-10-12 2022-09-16 연세대학교 산학협력단 Device, method and program for speech impairment evaluation
KR102511388B1 (en) * 2021-02-10 2023-03-17 휴브알엔씨 주식회사 A automatic analysis method of articulation disorder
KR102613014B1 (en) * 2021-08-12 2023-12-11 서울대학교산학협력단 Method for Automatic Severity Classification of Dysarthria

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
비특허문헌

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102064682B1 (en) 2018-09-21 2020-01-09 이화여자대학교 산학협력단 Method for fast mapping ability measurement using a quick incidental learning, recording medium and apparatus for performing the method
KR20200085041A (en) 2019-01-04 2020-07-14 순천향대학교 산학협력단 Language rehabilitation based vocal voice evaluation apparatus and method thereof
KR20210102620A (en) * 2020-02-12 2021-08-20 이화여자대학교 산학협력단 Video-based language disorder analysis system, method and program for performing the analysis
KR102336015B1 (en) * 2020-02-12 2021-12-06 이화여자대학교 산학협력단 Video-based language disorder analysis system, method and program for performing the analysis

Also Published As

Publication number Publication date
KR20170059664A (en) 2017-05-31

Similar Documents

Publication Publication Date Title
KR101804389B1 (en) System and method for autonomically testing dementia
Konig et al. Use of speech analyses within a mobile application for the assessment of cognitive impairment in elderly people
Dos Santos et al. A nonword repetition task to assess bilingual children’s phonology
Kittredge et al. Where is the effect of frequency in word production? Insights from aphasic picture-naming errors
AU2009299102B2 (en) Measuring cognitive load
Eadie et al. Profiles of grammatical morphology and sentence imitation in children with specific language impairment and Down syndrome
Lindsay et al. Language impairment in Alzheimer’s disease—robust and explainable evidence for ad-related deterioration of spontaneous speech through multilingual machine learning
Parish-Morris et al. Exploring autism spectrum disorders using HLT
Zimmerer et al. Automated profiling of spontaneous speech in primary progressive aphasia and behavioral-variant frontotemporal dementia: an approach based on usage-frequency
Hoffman et al. Concurrent and construct validity of oral language measures with school-age children with specific language impairment
Knilans et al. Online sentence reading in people with aphasia: Evidence from eye tracking
de Bree et al. Bempen or bemben: Differences between children at-risk of dyslexia and children with SLI on a morpho-phonological task
Castilla-Earls et al. The complexity of the Spanish subjunctive in bilingual children with SLI
Chaudhary et al. A systematic review on the role of language-related factors in the manifestation of stuttering in bilinguals
Shinkawa et al. Word repetition in separate conversations for detecting dementia: a preliminary evaluation on data of regular monitoring service
Carbonieri et al. Vocabulary assessment in Brazilian children: a systematic review with three instruments
Martínez‐Ferreiro Naming as a window to word retrieval changes in healthy and pathological ageing: Methodological considerations
Sheng et al. Developmental language disorder in Chinese children: A systematic review of research from 1997 to 2022
Yamada et al. A mobile application using automatic speech analysis for classifying Alzheimer's disease and mild cognitive impairment
Kacinik et al. Predicting noun and verb latencies: Influential variables and task effects
Ossewaarde et al. Quantifying the uncertainty of parameters measured in spontaneous speech of speakers with dementia
Jarrold et al. A" Verbal Thermometer" for Assessing Neurodegenerative Disease: Automated Measurement of Pronoun and Verb Ratio from Speech
Treviso et al. Detecting mild cognitive impairment in narratives in Brazilian Portuguese: first steps towards a fully automated system
Aluja et al. The dimensional assessment of personality psychopathology basic questionnaire: Shortened versions item analysis
Quené How to design and analyze language acquisition studies

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E701 Decision to grant or registration of patent right
GRNT Written decision to grant