CN109431522A - Autism early screening device based on name reaction normal form - Google Patents
Autism early screening device based on name reaction normal form Download PDFInfo
- Publication number
- CN109431522A CN109431522A CN201811220414.XA CN201811220414A CN109431522A CN 109431522 A CN109431522 A CN 109431522A CN 201811220414 A CN201811220414 A CN 201811220414A CN 109431522 A CN109431522 A CN 109431522A
- Authority
- CN
- China
- Prior art keywords
- face
- name
- reaction
- experimenter
- module
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/167—Personality evaluation
Abstract
The invention discloses a kind of autism early screening devices based on name reaction normal form, it include: acquisition module, for acquiring the voice signal for participating in staff's name, the picture signal of experimenter's behavioral activity, the experimenter of participation includes suffering from ASD individual and normal individual；Speech recognition module carries out the time point of name when name reaction test with positioning for identification；Face datection mark module, for the face and profile from face from extraction face characteristic in the initial data of video camera and use reference point identifying technical mark；Children's human face data is preserved for subsequent analysis by face detection module；Characteristic extracting module extracts response delay time, the duration of the reaction, reaction head pose feature of experimenter from the picture and mark tally evidence of the children's face detected；Whether prediction module is that ASD patient carries out assessment prediction to tester.The present invention more standardizes, and is more advantageous to automatization judgement and analysis, can automatically derive result.
Description
Technical field
The present invention relates to speech recognition, field of image processing more particularly to it is a kind of based on name reaction normal form it is lonely
Disease early screening device, acquire name person's name when audio and subject observation people by name when reaction image data and be subject to
Analysis, with the device of assessment prediction autism-spectrum obstacle.
Background technique
Society is big in the past few years to the concern of autism-spectrum obstacle (autism spectrum disorder, ASD)
Width rises, and just has 1 people in the U.S., every 68 people and suffers from autism.Although existing ASD appraisal procedure is highly effective,
It is time-consuming and laborious；And most of diagnostic methods mainly to communication obstacle, social handicap, repeat stereotypic behavior this three
The assessment of aspect.Name response analysis is one of the method being widely used in screening autism.This method needs one and faces
The veteran professional's operation experiments of bed, and substantially judged according to clinical experience.It is more that this not only needs diagnosis and treatment personnel to possess
The diagnostic experiences in year, non-quantized, nonstandardized technique judgment mode also limit accuracy of the autism by early screening.
Name response analysis, has studied how ASD individual responds called name.These research unanimously think ASD individual with
The child of normal growth compares, and often has different behavioural characteristics when by name.So far, most of researchs all rest on
It was found that the regular few people in statistical significance apply it on the early screening of autism.
The shortcomings that based on artificial name reaction normal form in practical applications is that it acts individual reaction time, response
The measurement and its difficulty of head angle.How systematic acquisition experimental data, and automated using machine learning method
Evaluation, become a challenging problem.
Summary of the invention
In view of the above technical problems, the purpose of the present invention is to provide a kind of autism early stages based on name reaction normal form
Screening apparatus, the present invention can standardize the collection process of name response data, predict the risk degree of ASD, screening ASD
Individual, auxiliary ASD diagnosis, improves the chance of ASD early prediction.
To achieve the above object, the present invention is realized according to following technical scheme:
A kind of autism early screening device based on name reaction normal form characterized by comprising
Acquisition module, for acquiring the voice signal for participating in staff's name, the image letter of experimenter's behavioral activity
Number, the experimenter of participation includes with ASD individual and normal individual；
Speech recognition module carries out the time point of name when name reaction test with positioning for identification；
Face datection mark module, for extracting face characteristic from the initial data of video camera and using reference point identifying
The face and profile of face at technical mark；
Face detection module for identifying from image data into face and children's face, and distinguishes, by children's face
Data are preserved for subsequent analysis；
Characteristic extracting module extracts the reaction of experimenter from the picture and mark tally evidence of the children's face detected
Delay time, duration of the reaction, reaction head pose feature；
Prediction module, the prediction module have using ASD label known in training data and the feature extracted, training
Supervised learning classifier, and class test is carried out to the feature extracted in test data, to tester whether be ASD patient into
Row assessment prediction.
In above-mentioned technical proposal, the response delay time is used to extract the name answering delay time of experimenter, root
It is calculated according to following formula:
Latency=Tf-Tc1,
Wherein TfIndicate that video camera detects the time of face, T for the first timec1Indicate the time that the first sound name starts.
In above-mentioned technical proposal, the duration of the reaction is used to extract the name response duration time of experimenter, root
It is calculated according to following formula:
Duration=Nf/ framerate,
Wherein, NfExpression detects the number of image frames of experimenter's face, and framerate each second, how many was for analyzing
Frame data.
In above-mentioned technical proposal, the reaction head pose feature is by n face key feature points (land marker)
Picture coordinate data, is calculated according to the following formula:
F=[x '1-x′2,…,x′1-x′n,x′2-x′3,…,x′n-1-x′n,
y′1-y′2,…,y1-y′n,y2-y′3,…,y′n-1-y′n]
Wherein xi,yiIt is the coordinate of i-th of key point, xi' and yi' normalization formula calculating according to the following formula:
Compared with the prior art, the invention has the following advantages:
The invention proposes the frame based on machine learning, the outer phenotypic characteristic of subordinate act sets out, when to subject by name
The different mode of responsing reaction is analyzed, and a kind of early screening device for predicting ASD is put forward.It is manually cried compared to traditional
Name normal form diagnostic method, method data proposed by the present invention more standardize, and are more advantageous to judgement and analysis, and can automatically derive
As a result.Although device proposed by the present invention can not substitute traditional ASD diagnostic method completely, an ASD can be considered and commented
The auxiliary early screening device estimated, so that the ASD assessment prediction of early stage is more accurately and conveniently.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the present invention prediction total frame construction drawing of ASD device；
The process schematic that device uses in Fig. 2 present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.
As shown in Figure 1, a kind of autism early screening device based on name reaction normal form of the invention, comprising:
Acquisition module, for acquiring the voice signal for participating in staff's name, the image letter of experimenter's behavioral activity
Number, the experimenter of participation includes with ASD individual and normal individual；
In the present embodiment, 43 people are tested, wherein 22 people are ASD children, 21 people are normal child.
Speech recognition module carries out the time point of name when name reaction test with positioning for identification；
Specifically, experimental field voice signal is acquired, the time point of staff's first time name is obtained.
Face datection mark module, for extracting face characteristic from the initial data of video camera and using reference point identifying
The face and profile of face at technical mark；
Face detection module for identifying from image data into face and children's face, and distinguishes, by children's face
Data are preserved for subsequent analysis；The human face data of non-subject children is ignored；
Characteristic extracting module extracts the reaction of experimenter from the picture and mark tally evidence of the children's face detected
Delay time, duration of the reaction, reaction head pose feature.
In the present embodiment, identify that the reaction head pose feature of experimenter and number of image frames carry out identification experimenter and be
The no time point for making responsing reaction and response and time span.
In above-mentioned technical proposal, the response delay time is used to extract the name answering delay time of experimenter, root
It is calculated according to following formula:
Latency=Tf-Tc1,
Wherein TfIndicate that video camera detects the time of face, T for the first timec1Indicate the time that the first sound name starts.
In above-mentioned technical proposal, the duration of the reaction is used to extract the name response duration time of experimenter, root
It is calculated according to following formula:
Duration=Nf/ framerate,
Wherein, NfExpression detects the number of image frames of experimenter's face, and framerate each second, how many was for analyzing
Frame data.
In above-mentioned technical proposal, the reaction head pose feature by n face key feature points coordinate data, according to
Lower formula is calculated:
F=[x '1-x′2,…,x′1-x′n,x′2-x′3,…,x′n-1-x′n,
y′1-y′2,…,y1-y′n,y2-y′3,…,y′n-1-y′n]
Wherein xi,yiIt is the coordinate of i-th of key point, xi' and yi' normalization formula calculating according to the following formula:
In this embodiment, in order to avoid excessively high-dimensional situation, principal component analysis PCA is applied to extract information content most
The data of several high dimensions.
The invention also includes prediction module, the prediction module is using ASD label known in training data and extracts
Feature, training supervised learning classifier, and carry out class test to the feature extracted in test data is to tester
It is no to carry out assessment prediction for ASD patient.
Specifically, by the data characteristics of tester, two classification analysis are carried out using logistic model, assessment will be by result
Display instrument is shown as suffering from the value-at-risk of autism.
In this embodiment, the confusion matrix of assessment result such as the following table 1:
Table 1
The accuracy rate of response delay time is 90.7%, and the accuracy rate for weighting response duration time is 90.7%, and name is anti-
The accuracy rate for answering the automatic assessment device of normal form is 93%.
Although above having used general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention are fallen within the scope of the claimed invention.
Claims (4)
1. a kind of autism early screening device based on name reaction normal form characterized by comprising
Acquisition module, for acquiring the voice signal for participating in staff's name, the picture signal of experimenter's behavioral activity, ginseng
With experimenter include with ASD individual and normal individual；
Speech recognition module carries out the time point of name when name reaction test with positioning for identification；
Face datection mark module, for extracting face characteristic from the initial data of video camera and using reference point identifying technology
The face and profile of mark face；
Face detection module for identifying from image data into face and children's face, and distinguishes, by children's human face data
It is preserved for subsequent analysis；
Characteristic extracting module extracts the response delay of experimenter from the picture and mark tally evidence of the children's face detected
Time, duration of the reaction, reaction head pose feature；
Prediction module trains supervised learning classifier using ASD label known in training data and the feature extracted,
And class test is carried out to the feature extracted in test data, it whether is that ASD patient carries out assessment prediction to tester.
2. the autism early screening device according to claim 1 based on name reaction normal form, which is characterized in that described
The response delay time is used to extract the name answering delay time of experimenter, calculates according to the following formula:
Latency=Tf-Tc1,
Wherein TfIndicate that video camera detects the time of face, T for the first timec1Indicate the time that the first sound name starts.
3. the autism early screening device according to claim 2 based on name reaction normal form, which is characterized in that described
Duration of the reaction is used to extract the name response duration time of experimenter, calculates according to the following formula:
Duration=Nf/ framerate,
Wherein, NfExpression detects the number of image frames of experimenter's face, and framerate is that each second, how many was used for analysis
Frame data.
4. the autism early screening device according to claim 3 based on name reaction normal form, which is characterized in that described
Head pose feature is reacted by n face key feature points coordinate data, is calculated according to the following formula:
F=[x '1-x′2,…,x′1-x′n,x′2-x′3,…,x′n-1-x′n,
y′1-y′2,…,y1-y′n,y2-y′3,…,y′n-1-y′n]
Wherein xi,yiIt is the coordinate of i-th of key point, xi' and yi' normalization formula calculating according to the following formula:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811220414.XA CN109431522A (en) | 2018-10-19 | 2018-10-19 | Autism early screening device based on name reaction normal form |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811220414.XA CN109431522A (en) | 2018-10-19 | 2018-10-19 | Autism early screening device based on name reaction normal form |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109431522A true CN109431522A (en) | 2019-03-08 |
Family
ID=65547500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811220414.XA Pending CN109431522A (en) | 2018-10-19 | 2018-10-19 | Autism early screening device based on name reaction normal form |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109431522A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110353703A (en) * | 2019-07-05 | 2019-10-22 | 昆山杜克大学 | Autism based on language paradigm behavioural analysis of repeating the words of others like a parrot assesses apparatus and system |
| CN110364260A (en) * | 2019-07-05 | 2019-10-22 | 昆山杜克大学 | Autism earlier evaluations apparatus and system based on indicative language paradigm |
| CN112163512A (en) * | 2020-09-25 | 2021-01-01 | 杨铠郗 | Autism spectrum disorder face screening method based on machine learning |
| CN113080964A (en) * | 2021-03-12 | 2021-07-09 | 广州市启路健康科技有限公司 | Self-closing data processing method and device based on intervention robot |
| CN114446476A (en) * | 2022-01-28 | 2022-05-06 | 中南大学湘雅二医院 | Construction method, prediction method and device of autism treatment effect prediction model |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069304A (en) * | 2015-08-18 | 2015-11-18 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Machine learning-based method for evaluating and predicting ASD |
| CN105279387A (en) * | 2015-11-17 | 2016-01-27 | 东南大学 | Execution function evaluating and training system for autism spectrum disorder children |
| CN108143411A (en) * | 2017-12-13 | 2018-06-12 | 东南大学 | A kind of tranquillization state brain electricity analytical system towards Autism Diagnostic |
-
2018
- 2018-10-19 CN CN201811220414.XA patent/CN109431522A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069304A (en) * | 2015-08-18 | 2015-11-18 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Machine learning-based method for evaluating and predicting ASD |
| CN105279387A (en) * | 2015-11-17 | 2016-01-27 | 东南大学 | Execution function evaluating and training system for autism spectrum disorder children |
| CN108143411A (en) * | 2017-12-13 | 2018-06-12 | 东南大学 | A kind of tranquillization state brain electricity analytical system towards Autism Diagnostic |
Non-Patent Citations (1)
| Title |
|---|
| WENBO LIU ET AL: "Response to name: A dataset and a multimodal machine learning framework towards autism study", 《2017 SEVENTH INTERNATIONAL CONFERENCE ON AFFECTIVE COMPUTING AND INTELLIGENT INTERACTION (ACII)》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110353703A (en) * | 2019-07-05 | 2019-10-22 | 昆山杜克大学 | Autism based on language paradigm behavioural analysis of repeating the words of others like a parrot assesses apparatus and system |
| CN110364260A (en) * | 2019-07-05 | 2019-10-22 | 昆山杜克大学 | Autism earlier evaluations apparatus and system based on indicative language paradigm |
| CN110353703B (en) * | 2019-07-05 | 2021-11-09 | 昆山杜克大学 | Autism assessment device and system based on parrot tongue learning language model behavior analysis |
| CN112163512A (en) * | 2020-09-25 | 2021-01-01 | 杨铠郗 | Autism spectrum disorder face screening method based on machine learning |
| CN113080964A (en) * | 2021-03-12 | 2021-07-09 | 广州市启路健康科技有限公司 | Self-closing data processing method and device based on intervention robot |
| CN114446476A (en) * | 2022-01-28 | 2022-05-06 | 中南大学湘雅二医院 | Construction method, prediction method and device of autism treatment effect prediction model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109431522A (en) | Autism early screening device based on name reaction normal form | |
| US11033221B2 (en) | Method and device for swallowing impairment detection | |
| CN113069080B (en) | Difficult airway assessment method and device based on artificial intelligence | |
| CN106725532A (en) | Depression automatic evaluation system and method based on phonetic feature and machine learning | |
| US10959661B2 (en) | Quantification of bulbar function | |
| CN108652587B (en) | Cognitive dysfunction prevention monitoring devices | |
| CN109381200B (en) | Primary screening device for autism based on separation and reunion normal form and computer vision | |
| CN114220543B (en) | Body and mind pain index evaluation method and system for tumor patient | |
| Simply et al. | Diagnosis of obstructive sleep apnea using speech signals from awake subjects | |
| CN103345633A (en) | Structural nuclear magnetic resonance image processing method used for Alzheimer disease early detection | |
| CN114565957A (en) | Consciousness assessment method and system based on micro expression recognition | |
| CN111128368B (en) | Automatic autism spectrum disorder detection method and device based on video expression behavior analysis | |
| US11222653B2 (en) | System and method for determining stroke based on voice analysis | |
| CN113329683A (en) | Method and device for screening for swallowing impairment | |
| CN110364260A (en) | Autism earlier evaluations apparatus and system based on indicative language paradigm | |
| WO2018073113A1 (en) | Device for determining features of a person | |
| CN111048202A (en) | Intelligent traditional Chinese medicine diagnosis system and method thereof | |
| Gomez et al. | Articulation acoustic kinematics in ALS speech | |
| CN113539486A (en) | Health state identification system based on traditional Chinese medicine facial and tongue manifestation dynamic change | |
| CN112507959A (en) | Method for establishing emotion perception model based on individual face analysis in video | |
| EP4276852A2 (en) | System for acoustic identification of obstruction types in sleep apnoea, and corresponding method | |
| CN110353703A (en) | Autism based on language paradigm behavioural analysis of repeating the words of others like a parrot assesses apparatus and system | |
| CN111460952B (en) | Method, system and prediction system for generating face recognition rule of schizophrenia | |
| Dubois et al. | Detecting and preventing falls with depth camera, tracking the body center | |
| CN109640819A (en) | For the asthma attack of test object or the equipment, system and method for asthma |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190308 |
|
| RJ01 | Rejection of invention patent application after publication |