KR20110060618A - Method and system for diagnosing handicapped person based on surveyed data - Google Patents

Abstract

PURPOSE: A method and a system for diagnosing disability are provided to diagnose disability based on recording information by utilizing a database for developmental disabilities using a database structure. CONSTITUTION: An information collection unit(610) collect information by observing developmental disabilities. An information extraction unit(620) selects a sentence by analyzing the collected information on the basis of natural language. The information extraction unit performs a morphological analysis process by performing the natural language process. A group generation unit(630) generates a group by extracting a correlation index for the selected word or the selected sentence. A disability diagnosis unit(640) classifies the disabilities for the developmental disabilities on the basis of the group and generates the diagnosed results.

Description

Disability Diagnosis Method and System Based on Observational Information on Developmental Disabilities {METHOD AND SYSTEM FOR DIAGNOSING HANDICAPPED PERSON BASED ON SURVEYED DATA}

The present invention discloses a method and system for diagnosing disability based on observational information on a developmental disability.

Recently, due to the development of social function and the change of the surrounding environment, the number of people with social development disorder tendency is increasing. In order to promote the social activities of people with this tendency, systems for storing and observing records of unusual behaviors in the form of databases for special education and rehabilitation from the perspective of special education and rehabilitation have been proposed. That is, in the past, the approach has been approached from a management point of view regarding the record / observation / education process for the developmental disabled. However, the conventionally proposed methods only approach the record side of the disabled, and thus, they cannot obtain more valuable information using the scaled database.

Currently, people with developmental disabilities are seeing a very wide range of autism tendencies (ASDs), and it is necessary to provide appropriate rehabilitation education and treatment for the developmental disabilities based on accurate disability diagnosis.

In order to solve the above problems, there is a need for a method for providing and diagnosing a disability diagnosis for a developmental disability using existing accumulated information.

The present invention proposes a system that can utilize a database for a developmentally disabled person, using an existing database or a new type of database structure, and perform a disability diagnosis based on record information for a developmentally disabled person.

In order to achieve the above object, a disability diagnosis method based on observation of a developmental disability according to an embodiment of the present invention is a first step of collecting information by observing a developmental disability, and analyzing the collected information based on natural language A second step of selecting a word corresponding to the observation result or a sentence including two or more words, a third step of generating a cluster by extracting a correlation index with respect to the selected word or sentence, and based on the generated cluster And classifying a disability for the developmental disability and generating a diagnosis result.

An information collecting unit for collecting information by observing a developmental disability according to another embodiment of the present invention, analyzing the collected information based on natural language to select a word or a sentence including two or more words corresponding to the observation result An information extraction unit, a cluster generation unit for generating a cluster by extracting a correlation index with respect to the selected word or sentence, and a disorder diagnosis for classifying the disorder for the developmental disabled based on the generated cluster and generating a diagnosis result Contains wealth.

The present invention allows to present the results more objectively by using a daily life records DB that has been established in large capacity to diagnose a developmental disability.

In addition, the present invention can provide an opportunity to discover new phenomena and values in the existing massively built database, and can be used as an objective approach tool in diagnosing disorders in the future.

In addition, since the brain function analysis data can be used to apply more objective brain dysfunction differences, it is possible to suggest a diagnosis technique based on actual biological responses.

Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings. In adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are assigned to the same components as much as possible even though they are shown in different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In addition, in describing the component of this invention, terms, such as 1st, 2nd, A, B, (a), (b), can be used. These terms are only for distinguishing the components from other components, and the nature, order or order of the components are not limited by the terms. If a component is described as being "connected", "coupled" or "connected" to another component, that component may be directly connected or connected to that other component, but between components It will be understood that may be "connected", "coupled" or "connected".

In one embodiment of the present invention, cognitive function, language function, social function, motor function, daily function for attention deficit hyperactivity disorder (ADHD) and developmental disorders (or children with developmental disorders) with autism, epilepsy, etc. In order to more accurately diagnose the disability of people who have problems with life functions, we propose a technique for diagnosing disability based on the records of daily life characteristics and special behaviors in the field of special education and rehabilitation.

1 is a diagram illustrating a disability diagnosis method based on observation information on a developmental disability according to an embodiment of the present invention. Each component of FIG. 1 may be one device or system, and may be a component constituting the device or system. That is, it may be an element constituting one system, or may constitute a plurality of systems connected to a network to exchange information.

Raw data for each person with a developmental disability are selected (130). Recorded data can be data describing everyday life, and can be data describing a particular behavior. Then, text preprocessing is performed on the selected data (140). The pretreatment according to an embodiment of the present invention may be processed using a morpheme analyzer, which is a base technology of the existing natural language processing technology. To this end, tagging may be performed on a specific word, which may be a noun, a verb, etc., based on the specific word. Existing morphological analyzer can be used to extract features for fault diagnosis. Features in one embodiment of the present invention includes specifics about disability for each function (cognition, language, sociality, athletic ability, daily life), word (word and verb) based, phrase based, sentence Based on the specific pattern can be configured. In addition, the feature extraction can be extracted based on the preceding dictionary information 110 for unusual behavior, and it is composed of a verb (noun, rhetoric, pronoun, etc.) and a verb (verb, adjective, etc.) without prior information on the unusual behavior. Extract as a feature.

The singular behavior record DB 110 includes information on the specific behavior and may include classification information on the specific behavior. For example, it can function as a database by storing information or classification information about specific behaviors such as medical, interventional, observational, and therapist's positive specific behaviors. Of course, the information generated while performing 140 may be stored in the specific behavior record DB 110 again. The singular behavior recording DB 110 includes conventionally recorded information.

A correlation index based on the feature word is extracted based on the extracted feature, and clustering is performed (150). In an embodiment of the present invention, clustering may be performed by extracting a correlation index based on a weight and a frequency number on a time domain. In the process of extracting and clustering the correlation index at 150, the specific behavior record DB 110 may be used, and newly derived information may be stored in the specific behavior record DB 110.

Based on the extracted information, the failure is classified and a diagnosis result is derived (160). In this process, the failure classification reference DB 120 may be referred to, and the failure classification and diagnosis result information may be stored in the failure classification reference DB 120 again. The information can be used for measuring functional near-infrared spectroscopy (fNIRS), which is behavior detection based on brain signals (170).

Finally, at 170, fNIRS-based brain function data can be analyzed and analyzed to verify brain activity similarity in order to provide a more general and objective form of disability diagnosis. This data is used to further update the failure classification criteria database to improve gradual accuracy in fault diagnosis.

2 is a diagram illustrating a text preprocessing and a correlation index extraction and clustering process according to an embodiment of the present invention.

Previously, the result of performing morphological analysis based on the selected data by selecting the record data (Raw-data) for each developmentally disabled person is shown as "field 2" of FIG. Here, the statement uses what is tagged as "* / N" (indicated by a red square) as feature information. Of course, the statement includes nouns, pronouns, rhetoric, and the like. Other features include adjectives and verbs in using verbs.

2 is a schematic diagram illustrating one processing procedure. The frequency of each feature, freq (term or verb), can be calculated as in Equation 1.

[Equation 1]

In addition, by using the time-based record information of the developmental disability, f (speak or verb, record_start time, most recent_record_time) information for a specific period can be calculated as shown in Equation 2.

[Equation 2]

Equation 2 calculates the frequency of consecutive f (N) in a predetermined time interval (time interval between t_s and t_e), where N can be spoken. In addition, Total_freq means the frequency of N in the whole treatment information, Local_freq can be a formula for calculating the frequency for children with developmental disabilities.

Finally, the final correlation index is calculated using Equation 3 using time-based record information on the features (eg, verbs) included for each person with developmental disabilities.

[Equation 3]

In Equation 3, the frequency of consecutive Ns in a predetermined time interval (time interval between t_s and t_e) is obtained, where N is the total number of extracted messages for a child with developmental disabilities. The speech list may be {stool, cold symptoms, irregular}, and the speech list may be composed of {specific situation, annoying, normal, defecation, irregular, stool} indicated by a red square in FIG. 2.

3 is a result of finally calculating the correlation index calculated through Equation 3 for each developmentally disabled person. Here, it can be used as a basic data for evaluating disability diagnosis by clustering by the developmental disabilities having a small difference in the correlation index for each developmental disability. In evaluating the diagnosis of disability, the final classification criteria are determined based on the prior classification criteria and the new classification criteria.

4 is a diagram illustrating an example of performing one clustering by graphically representing the data of FIG. 3. As shown in FIG. 4, the smaller the difference in the correlation index is, the higher the likelihood that a diagnosis of the same disorder is made. 4 shows clustered values for Normal, attention deficit hyperactivity disorder (ADHD), or Austism.

5 is a diagram illustrating a disability diagnosis process based on observation information on a developmental disability according to an embodiment of the present invention. First, the information is collected by observing a person with a developmental disability as listed in the specific behavior region of FIG. 2 (S510). Collection of information means collecting information necessary to extract feature information. The collected information is analyzed based on natural language to select a word corresponding to an observation result or a sentence including two or more words (S520). Selection of a word or sentence means selecting a word to be analyzed in field 2 of FIG. 2.

A cluster is generated by extracting a correlation index with respect to the selected word or sentence (S530). One embodiment of the process of obtaining a correlation index includes a method of calculating a frequency of occurrence of a specific word or sentence within a predetermined time interval as described in Equations 1, 2, and 3. Then, based on the generated cluster, the disorder for the developmental disability is classified and a diagnosis result is generated (S540). An example of the generated cluster is shown in FIG. 4. Through a cluster as shown in FIG. 4, a diagnosis result may be generated by comparing characteristics of a disorder corresponding to a predetermined cluster (cluster) characteristic. For example, if the frequency of annoying behavior is very high, it can be categorized as a disorder in which annoyance is a major feature.

Referring to the process of FIG. 5 in more detail, the process S520 may proceed with morphological analysis by natural language processing. This means a process of extracting a statement or a verb including information on the state or behavior of a developmental disability by performing morphological analysis on a specific word or sentence as shown in FIG. 2. Also, for more accurate information, the representative feature words for the selected words or sentences can be selected in the S520 process, which is the same as the representative information from the collected information, but is clearly selected by selecting a different feature as one representative feature. This is to ensure feature information.

For example, if the information is recorded as 'produce nervous' and 'provoke', the two acts have similarities with each other and can be calculated as the representative feature, 'provoke'. As another example, a cluster may be created by selecting a representative feature of 'cry' for similar behaviors such as 'cry', 'crying', and 'jingling'. Of course, it is also possible to substitute and represent the representative feature words in the process of collecting information.

As described above, the process of calculating the frequency may calculate the frequency of occurrence of the developmental disability and the total developmental disability for the representative feature word or the selected word / phrase. As an example of the calculation process has been described in Equation 1, 2, 3. For example, if the word 'squeeze' occurs five times within a three-hour interval in the information collected about autistic developmental disabilities, and four times 'squeeze' occurs in the collected information on a specific developmental disability, Can be included in a community of people with autism.

In the process of FIG. 5, in order to select a word or sentence, a database storing information on a state or behavior of a developmental disabled may be searched. In FIG. 1, in the process of measuring the text preprocessing and the correlation index, the process of referring to and updating the specific behavior record DB may be performed.

In S540, the disorder for the developmental disability is classified and a diagnosis result is generated, and based on the result, it may be used for brain signal based behavior detection (fNIRS) measurement.

6 is a diagram illustrating a configuration of a disability diagnosis system based on observation information on a developmental disability according to an embodiment of the present invention. In general, the information collection unit 610, the information extraction unit 620, the cluster generation unit 630, and comprises a failure diagnosis unit 640.

The information collecting unit 610 collects information by observing a developmental disability, and the information extracting unit 620 analyzes the collected information based on natural language, and includes a word corresponding to an observation result or a sentence including two or more words. Screening.

The cluster generator 630 generates a cluster by extracting a correlation index for the selected word or sentence, and the disorder diagnosis unit 640 classifies and diagnoses a disorder for the developmental disability based on the generated cluster. Produces a result.

In more detail, the information extracting unit 620 may perform morphological analysis by natural language processing and select a representative feature for a selected word or sentence. Representative feature words will be replaced by the above description.

The cluster generation unit 630 calculates a frequency occurring in all developmentally disabled persons and a frequency occurring in the developmentally disabled person for a specific period with respect to the selected representative feature word, and generates a cluster of the representative feature words using the calculated frequency. can do.

In the structure of FIG. 6, a word corresponding to an observation result may be a statement or a verb including information on a state or behavior of a developmental disabled person. The information extracting unit 620 may search a database that stores information about a state or behavior of a person with a developmental disability. Information calculated by the system of FIG. 6, that is, the diagnosis result generated by the disorder diagnosis unit 640 may be used as basic information for measuring brain signal based behavior detection (fNIRS).

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention but to describe the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

1 is a diagram illustrating a disability diagnosis method based on observation information on a developmental disability according to an embodiment of the present invention.

2 is a diagram illustrating a text preprocessing and a correlation index extraction and clustering process according to an embodiment of the present invention.

3 is a result of finally calculating the correlation index calculated through Equation 3 for each developmentally disabled person.

4 is a diagram illustrating an example of performing one clustering by graphically representing the data of FIG. 3.

5 is a diagram illustrating a disability diagnosis process based on observation information on a developmental disability according to an embodiment of the present invention.

6 is a diagram illustrating a configuration of a disability diagnosis system based on observation information on a developmental disability according to an embodiment of the present invention.

Claims (12)

A first step of collecting information by observing a developmental disability; A second step of analyzing the collected information based on natural language and selecting a word corresponding to an observation result or a sentence including two or more words; Generating a cluster by extracting a correlation index with respect to the selected word or sentence; And And a fourth step of classifying the disability for the developmental disability based on the generated cluster, and generating a diagnosis result. The method of claim 1, Wherein the second step further comprises the step of proceeding morphological analysis by natural language processing, disability diagnosis method based on observation of developmental disabilities. The method of claim 1, The second step further includes selecting a representative feature for the selected word or sentence, The third step may include calculating a frequency occurring in all developmentally disabled persons and a frequency occurring in the developmentally disabled person for a selected period with respect to the selected representative feature word; And And generating a cluster of the representative feature word using the calculated frequency. The method of claim 1, The word corresponding to the observation result is a disability diagnosis method based on observation information on the developmental disability, characterized in that it includes a statement or a verb containing information about the state or behavior of the developmental disabled. The method of claim 1, The selecting step includes searching for a database storing information on a state or behavior of a developmental disabled, disability diagnosis based on observation of developmental disabilities. The method of claim 1, And the diagnosis result generated in the fourth step is used as basic information for measuring brain signal based behavior detection (fNIRS). An information collection unit for collecting information by observing a developmental disability; An information extraction unit for analyzing the collected information based on natural language and selecting a word corresponding to an observation result or a sentence including two or more words; And A cluster generation unit for generating a cluster by extracting a correlation index with respect to the selected word or sentence; And And a disability diagnosis unit configured to classify the disability for the developmental disability based on the generated cluster and generate a diagnosis result. The method of claim 7, wherein The information extraction unit is characterized in that the morphological analysis by processing the natural language, disability diagnosis system based on the observation of developmental disabilities. The method of claim 7, wherein The information extracting unit selects a representative feature word for the selected word or sentence, The cluster generation unit calculates a frequency occurring in all developmentally disabled persons and a frequency occurring in the developmentally disabled person for a specific period with respect to the selected representative feature word, and generates a cluster of the representative feature words using the calculated frequency. Disability diagnosis system based on observation of developmental disabilities. The method of claim 7, wherein The word corresponding to the observation result is a disability diagnosis system based on the developmental disability observation information, characterized in that it includes a statement or a verb containing information about the status or behavior of the developmental disabled. The method of claim 7, wherein The information extracting unit is a disability diagnosis system based on observation of a developmental disability, characterized in that for searching a database storing information about the state or behavior of a developmental disability. The method of claim 7, wherein And a diagnosis result generated by the disability diagnosis unit is used as basic information for measuring brain signal based behavior detection (fNIRS).

Images