Testing - National Autism Resource Centre (NARC)

2019

Abdullah, A A; Rijal, S; Dash, S R

Evaluation on Machine Learning Algorithms for Classification of Autism Spectrum Disorder (ASD) Conference

1372 (1), Institute of Physics Publishing, 2019, ISSN: 17426588, (cited By 0).

Abstract | Links | BibTeX | Tags: Autism Spectrum Disorders, Behaviour Evaluations, Biomedical Engineering, Brain Mapping, Classification (of information), Decision Trees, Diseases, Feature Extraction, Feature Selection Methods, K Fold Cross Validations, Learning, Least Absolute Shrinkage and Selection Operators, Least Squares Approximations, Logistic Regressions, Machine Learning, Machine Learning Methods, Magnetic Resonance Imaging, Nearest Neighbor Search, Regression Analysis, Supervised Learning, Supervised Machine Learning

@conference{Abdullah2019,
title = {Evaluation on Machine Learning Algorithms for Classification of Autism Spectrum Disorder (ASD)},
author = {A A Abdullah and S Rijal and S R Dash},
editor = {Rahim Mustafa Zaaba Norali Noor S B A N B S K A N B A B M Fook C.Y. Yazid H.B.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076493636&doi=10.1088%2f1742-6596%2f1372%2f1%2f012052&partnerID=40&md5=2ec1bd9f6cf1e3afe965cc9e3792f536},
doi = {10.1088/1742-6596/1372/1/012052},
issn = {17426588},
year = {2019},
date = {2019-01-01},
journal = {Journal of Physics: Conference Series},
volume = {1372},
number = {1},
publisher = {Institute of Physics Publishing},
abstract = {Autism Spectrum Disorder (ASD) was characterized by delay in social interactions development, repetitive behaviors and narrow interest, which usually diagnosed with standard diagnostic tools such as Autism Diagnostic Observation Schedule (ADOS) and Autism Diagnostic Interview-Revised (ADIR-R). Previous work has implemented machine-learning methods for the classification of ASD, however they used different types of dataset such as brain images for MRI and EEG, risk genes in genetic profiles and behavior evaluation based on ADOS and ADI-R. Here a trial on using Autism Spectrum Questions (AQ) to build models that have higher potential to classify ASD was developed. In this research, Chi-square and Least Absolute Shrinkage and Selection Operator (LASSO) have been selected as feature selection methods to select the most important features for 3 supervised machine learning algorithms, which are Random Forest, Logistic Regression and K-Nearest Neighbors with K-fold cross validation. The performance was evaluated in which results Logistic Regression scored the highest accuracy with 97.541% using model with 13 selected features based on Chi-square selection method. © 2019 IOP Publishing Ltd. All rights reserved.},
note = {cited By 0},
keywords = {Autism Spectrum Disorders, Behaviour Evaluations, Biomedical Engineering, Brain Mapping, Classification (of information), Decision Trees, Diseases, Feature Extraction, Feature Selection Methods, K Fold Cross Validations, Learning, Least Absolute Shrinkage and Selection Operators, Least Squares Approximations, Logistic Regressions, Machine Learning, Machine Learning Methods, Magnetic Resonance Imaging, Nearest Neighbor Search, Regression Analysis, Supervised Learning, Supervised Machine Learning},
pubstate = {published},
tppubtype = {conference}
}

2018

Hariharan, M; Sindhu, R; Vijean, V; Yazid, H; Nadarajaw, T; Yaacob, S; Polat, K

Improved binary dragonfly optimization algorithm and wavelet packet based non-linear features for infant cry classification Journal Article

Computer Methods and Programs in Biomedicine, 155 , pp. 39-51, 2018, ISSN: 01692607, (cited By 21).

Abstract | Links | BibTeX | Tags: Accidents, Algorithms, Article, Artificial Neural Network, Asphyxia, Binary Dragonfly Optimization Aalgorithm, Classification (of information), Classification Algorithm, Classifier, Coding, Computer-Assisted, Constants and Coefficients, Crying, Database Systems, Databases, Deafness, Diagnosis, Energy, Entropy, Extraction, Extreme Learning Machine, Factual, Factual Database, Feature Extraction, Feature Selection Methods, Fuzzy System, Hearing Impairment, Human, Hunger, Infant, Infant Cry, Infant Cry Classifications, Jaundice, Kernel Method, Learning, Linear Predictive Coding, Machine Learning, Mathematical Transformations, Mel Frequency Cepstral Coefficient, Mel Frequency Cepstral Coefficients, Multi-Class Classification, Neural Networks, Nonlinear Dynamics, Nonlinear System, Optimization, Pain, Pathophysiology, Prematurity, Reproducibility, Reproducibility of Results, Signal Processing, Speech Recognition, Wavelet Analysis, Wavelet Packet, Wavelet Packet Transforms

@article{Hariharan201839,
title = {Improved binary dragonfly optimization algorithm and wavelet packet based non-linear features for infant cry classification},
author = {M Hariharan and R Sindhu and V Vijean and H Yazid and T Nadarajaw and S Yaacob and K Polat},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85036611215&doi=10.1016%2fj.cmpb.2017.11.021&partnerID=40&md5=1f3b17817b00f07cadad6eb61c0f4bf9},
doi = {10.1016/j.cmpb.2017.11.021},
issn = {01692607},
year = {2018},
date = {2018-01-01},
journal = {Computer Methods and Programs in Biomedicine},
volume = {155},
pages = {39-51},
publisher = {Elsevier Ireland Ltd},
abstract = {Background and objective Infant cry signal carries several levels of information about the reason for crying (hunger, pain, sleepiness and discomfort) or the pathological status (asphyxia, deaf, jaundice, premature condition and autism, etc.) of an infant and therefore suited for early diagnosis. In this work, combination of wavelet packet based features and Improved Binary Dragonfly Optimization based feature selection method was proposed to classify the different types of infant cry signals. Methods Cry signals from 2 different databases were utilized. First database contains 507 cry samples of normal (N), 340 cry samples of asphyxia (A), 879 cry samples of deaf (D), 350 cry samples of hungry (H) and 192 cry samples of pain (P). Second database contains 513 cry samples of jaundice (J), 531 samples of premature (Prem) and 45 samples of normal (N). Wavelet packet transform based energy and non-linear entropies (496 features), Linear Predictive Coding (LPC) based cepstral features (56 features), Mel-frequency Cepstral Coefficients (MFCCs) were extracted (16 features). The combined feature set consists of 568 features. To overcome the curse of dimensionality issue, improved binary dragonfly optimization algorithm (IBDFO) was proposed to select the most salient attributes or features. Finally, Extreme Learning Machine (ELM) kernel classifier was used to classify the different types of infant cry signals using all the features and highly informative features as well. Results Several experiments of two-class and multi-class classification of cry signals were conducted. In binary or two-class experiments, maximum accuracy of 90.18% for H Vs P, 100% for A Vs N, 100% for D Vs N and 97.61% J Vs Prem was achieved using the features selected (only 204 features out of 568) by IBDFO. For the classification of multiple cry signals (multi-class problem), the selected features could differentiate between three classes (N, A & D) with the accuracy of 100% and seven classes with the accuracy of 97.62%. Conclusion The experimental results indicated that the proposed combination of feature extraction and selection method offers suitable classification accuracy and may be employed to detect the subtle changes in the cry signals. © 2017 Elsevier B.V.},
note = {cited By 21},
keywords = {Accidents, Algorithms, Article, Artificial Neural Network, Asphyxia, Binary Dragonfly Optimization Aalgorithm, Classification (of information), Classification Algorithm, Classifier, Coding, Computer-Assisted, Constants and Coefficients, Crying, Database Systems, Databases, Deafness, Diagnosis, Energy, Entropy, Extraction, Extreme Learning Machine, Factual, Factual Database, Feature Extraction, Feature Selection Methods, Fuzzy System, Hearing Impairment, Human, Hunger, Infant, Infant Cry, Infant Cry Classifications, Jaundice, Kernel Method, Learning, Linear Predictive Coding, Machine Learning, Mathematical Transformations, Mel Frequency Cepstral Coefficient, Mel Frequency Cepstral Coefficients, Multi-Class Classification, Neural Networks, Nonlinear Dynamics, Nonlinear System, Optimization, Pain, Pathophysiology, Prematurity, Reproducibility, Reproducibility of Results, Signal Processing, Speech Recognition, Wavelet Analysis, Wavelet Packet, Wavelet Packet Transforms},
pubstate = {published},
tppubtype = {article}
}

Background and objective Infant cry signal carries several levels of information about the reason for crying (hunger, pain, sleepiness and discomfort) or the pathological status (asphyxia, deaf, jaundice, premature condition and autism, etc.) of an infant and therefore suited for early diagnosis. In this work, combination of wavelet packet based features and Improved Binary Dragonfly Optimization based feature selection method was proposed to classify the different types of infant cry signals. Methods Cry signals from 2 different databases were utilized. First database contains 507 cry samples of normal (N), 340 cry samples of asphyxia (A), 879 cry samples of deaf (D), 350 cry samples of hungry (H) and 192 cry samples of pain (P). Second database contains 513 cry samples of jaundice (J), 531 samples of premature (Prem) and 45 samples of normal (N). Wavelet packet transform based energy and non-linear entropies (496 features), Linear Predictive Coding (LPC) based cepstral features (56 features), Mel-frequency Cepstral Coefficients (MFCCs) were extracted (16 features). The combined feature set consists of 568 features. To overcome the curse of dimensionality issue, improved binary dragonfly optimization algorithm (IBDFO) was proposed to select the most salient attributes or features. Finally, Extreme Learning Machine (ELM) kernel classifier was used to classify the different types of infant cry signals using all the features and highly informative features as well. Results Several experiments of two-class and multi-class classification of cry signals were conducted. In binary or two-class experiments, maximum accuracy of 90.18% for H Vs P, 100% for A Vs N, 100% for D Vs N and 97.61% J Vs Prem was achieved using the features selected (only 204 features out of 568) by IBDFO. For the classification of multiple cry signals (multi-class problem), the selected features could differentiate between three classes (N, A & D) with the accuracy of 100% and seven classes with the accuracy of 97.62%. Conclusion The experimental results indicated that the proposed combination of feature extraction and selection method offers suitable classification accuracy and may be employed to detect the subtle changes in the cry signals. © 2017 Elsevier B.V.