2019 |
Misman, M F; Samah, A A; Ezudin, F A; Majid, H A; Shah, Z A; Hashim, H; Harun, M F Classification of adults with autism spectrum disorder using deep neural network Conference Institute of Electrical and Electronics Engineers Inc., 2019, ISBN: 9781728130415, (cited By 0). Abstract | Links | BibTeX | Tags: Autism Spectrum Disorders, Brain Disorders, Classification (of information), Classification Accuracy, Classification Methods, Clinical Tests, Cognitive Skill, Computer Aided Diagnosis, Deep Learning, Deep Neural Networks, Diseases, Learning, Machine Learning Methods, Screening Data, Support Vector Machines @conference{Misman201929, title = {Classification of adults with autism spectrum disorder using deep neural network}, author = {M F Misman and A A Samah and F A Ezudin and H A Majid and Z A Shah and H Hashim and M F Harun}, url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079349811&doi=10.1109%2fAiDAS47888.2019.8970823&partnerID=40&md5=dd727e950667359680a6dbcc4855422f}, doi = {10.1109/AiDAS47888.2019.8970823}, isbn = {9781728130415}, year = {2019}, date = {2019-01-01}, journal = {Proceedings - 2019 1st International Conference on Artificial Intelligence and Data Sciences, AiDAS 2019}, pages = {29-34}, publisher = {Institute of Electrical and Electronics Engineers Inc.}, abstract = {Autism Spectrum Disorder (ASD) is a developmental brain disorder that causes deficits in linguistic, communicative, and cognitive skills as well as social skills. Various application of Machine Learning has been applied apart from the clinical tests available, which has increased the performance in the diagnosis of this disorder. In this study, we applied the Deep Neural Network (DNN) architecture, which has been a popular method in recent years and proved to improve classification accuracy. This study aims to analyse the performance of DNN model in the diagnosis of ASD in terms of classification accuracy by using two datasets of adult ASD screening data. The results are then compared with the previous Machine Learning method by another researcher, which is Support Vector Machine (SVM). The accuracy achieved by the DNN model in the classification of ASD diagnosis is 99.40% on the first dataset and achieved 96.08% on the second dataset. Meanwhile, the SVM model achieved an accuracy of 95.24% and 95.08% using the first and second data, respectively. The results show that ASD cases can be accurately identified by implementing the DNN classification method using ASD adult screening data. © 2019 IEEE.}, note = {cited By 0}, keywords = {Autism Spectrum Disorders, Brain Disorders, Classification (of information), Classification Accuracy, Classification Methods, Clinical Tests, Cognitive Skill, Computer Aided Diagnosis, Deep Learning, Deep Neural Networks, Diseases, Learning, Machine Learning Methods, Screening Data, Support Vector Machines}, pubstate = {published}, tppubtype = {conference} } Autism Spectrum Disorder (ASD) is a developmental brain disorder that causes deficits in linguistic, communicative, and cognitive skills as well as social skills. Various application of Machine Learning has been applied apart from the clinical tests available, which has increased the performance in the diagnosis of this disorder. In this study, we applied the Deep Neural Network (DNN) architecture, which has been a popular method in recent years and proved to improve classification accuracy. This study aims to analyse the performance of DNN model in the diagnosis of ASD in terms of classification accuracy by using two datasets of adult ASD screening data. The results are then compared with the previous Machine Learning method by another researcher, which is Support Vector Machine (SVM). The accuracy achieved by the DNN model in the classification of ASD diagnosis is 99.40% on the first dataset and achieved 96.08% on the second dataset. Meanwhile, the SVM model achieved an accuracy of 95.24% and 95.08% using the first and second data, respectively. The results show that ASD cases can be accurately identified by implementing the DNN classification method using ASD adult screening data. © 2019 IEEE. |
2013 |
Selvaraj, J; Murugappan, M; Wan, K; Yaacob, S Classification of emotional states from electrocardiogram signals: A non-linear approach based on hurst Journal Article BioMedical Engineering Online, 12 (1), 2013, ISSN: 1475925X, (cited By 42). Abstract | Links | BibTeX | Tags: Adolescent, Adult, Aged, Article, Audio-Visual Stimulus, Autonomous Nervous Systems, Children, Classification Accuracy, Computer Based Training, Computer-Assisted, Electrocardiogram Signal, Electrocardiography, Emotion, Female, Fuzzy K-nearest Neighbor, Higher-Order Statistic (HOS), Human, Intellectual Disability, Interactive Computer Systems, Methodology, Middle Aged, Nonlinear Dynamics, Nonlinear System, Procedures, Real Time Systems, Signal Processing, Statistics, Young Adult @article{Selvaraj2013, title = {Classification of emotional states from electrocardiogram signals: A non-linear approach based on hurst}, author = {J Selvaraj and M Murugappan and K Wan and S Yaacob}, url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84879017985&doi=10.1186%2f1475-925X-12-44&partnerID=40&md5=18c5309ac9f3017f455480f1ff732a30}, doi = {10.1186/1475-925X-12-44}, issn = {1475925X}, year = {2013}, date = {2013-01-01}, journal = {BioMedical Engineering Online}, volume = {12}, number = {1}, publisher = {BioMed Central Ltd.}, abstract = {Background: Identifying the emotional state is helpful in applications involving patients with autism and other intellectual disabilities; computer-based training, human computer interaction etc. Electrocardiogram (ECG) signals, being an activity of the autonomous nervous system (ANS), reflect the underlying true emotional state of a person. However, the performance of various methods developed so far lacks accuracy, and more robust methods need to be developed to identify the emotional pattern associated with ECG signals.Methods: Emotional ECG data was obtained from sixty participants by inducing the six basic emotional states (happiness, sadness, fear, disgust, surprise and neutral) using audio-visual stimuli. The non-linear feature 'Hurst' was computed using Rescaled Range Statistics (RRS) and Finite Variance Scaling (FVS) methods. New Hurst features were proposed by combining the existing RRS and FVS methods with Higher Order Statistics (HOS). The features were then classified using four classifiers - Bayesian Classifier, Regression Tree, K- nearest neighbor and Fuzzy K-nearest neighbor. Seventy percent of the features were used for training and thirty percent for testing the algorithm.Results: Analysis of Variance (ANOVA) conveyed that Hurst and the proposed features were statistically significant (p < 0.001). Hurst computed using RRS and FVS methods showed similar classification accuracy. The features obtained by combining FVS and HOS performed better with a maximum accuracy of 92.87% and 76.45% for classifying the six emotional states using random and subject independent validation respectively.Conclusions: The results indicate that the combination of non-linear analysis and HOS tend to capture the finer emotional changes that can be seen in healthy ECG data. This work can be further fine tuned to develop a real time system. © 2013 Selvaraj et al.; licensee BioMed Central Ltd.}, note = {cited By 42}, keywords = {Adolescent, Adult, Aged, Article, Audio-Visual Stimulus, Autonomous Nervous Systems, Children, Classification Accuracy, Computer Based Training, Computer-Assisted, Electrocardiogram Signal, Electrocardiography, Emotion, Female, Fuzzy K-nearest Neighbor, Higher-Order Statistic (HOS), Human, Intellectual Disability, Interactive Computer Systems, Methodology, Middle Aged, Nonlinear Dynamics, Nonlinear System, Procedures, Real Time Systems, Signal Processing, Statistics, Young Adult}, pubstate = {published}, tppubtype = {article} } Background: Identifying the emotional state is helpful in applications involving patients with autism and other intellectual disabilities; computer-based training, human computer interaction etc. Electrocardiogram (ECG) signals, being an activity of the autonomous nervous system (ANS), reflect the underlying true emotional state of a person. However, the performance of various methods developed so far lacks accuracy, and more robust methods need to be developed to identify the emotional pattern associated with ECG signals.Methods: Emotional ECG data was obtained from sixty participants by inducing the six basic emotional states (happiness, sadness, fear, disgust, surprise and neutral) using audio-visual stimuli. The non-linear feature 'Hurst' was computed using Rescaled Range Statistics (RRS) and Finite Variance Scaling (FVS) methods. New Hurst features were proposed by combining the existing RRS and FVS methods with Higher Order Statistics (HOS). The features were then classified using four classifiers - Bayesian Classifier, Regression Tree, K- nearest neighbor and Fuzzy K-nearest neighbor. Seventy percent of the features were used for training and thirty percent for testing the algorithm.Results: Analysis of Variance (ANOVA) conveyed that Hurst and the proposed features were statistically significant (p < 0.001). Hurst computed using RRS and FVS methods showed similar classification accuracy. The features obtained by combining FVS and HOS performed better with a maximum accuracy of 92.87% and 76.45% for classifying the six emotional states using random and subject independent validation respectively.Conclusions: The results indicate that the combination of non-linear analysis and HOS tend to capture the finer emotional changes that can be seen in healthy ECG data. This work can be further fine tuned to develop a real time system. © 2013 Selvaraj et al.; licensee BioMed Central Ltd. |
2012 |
Hoole, P R P; Pirapaharan, K; Basar, S A; Ismail, R; Liyanage, D L D A; Senanayake, S S H M U L; Hoole, S R H Autism, EEG and brain electromagnetics research Conference 2012, ISBN: 9781467316668, (cited By 11). Abstract | Links | BibTeX | Tags: Biomedical Engineering, Brain, Brain Regions, Classification Accuracy, Diseases, EEG Signals, Electromagnetic Signals, Electromagnetics, Electromagnetism, Frequency Domains, International Group, Multilayer Perception Neural Networks, Neuroimaging, Principal Component Analysis @conference{Hoole2012541, title = {Autism, EEG and brain electromagnetics research}, author = {P R P Hoole and K Pirapaharan and S A Basar and R Ismail and D L D A Liyanage and S S H M U L Senanayake and S R H Hoole}, url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84876771339&doi=10.1109%2fIECBES.2012.6498036&partnerID=40&md5=9f9390b30b859a90936c66699c1a5115}, doi = {10.1109/IECBES.2012.6498036}, isbn = {9781467316668}, year = {2012}, date = {2012-01-01}, journal = {2012 IEEE-EMBS Conference on Biomedical Engineering and Sciences, IECBES 2012}, pages = {541-543}, abstract = {There has been a significant increase in the incidence of autism. We report the work on autism by our international group, on the growing attention paid to EEG based diagnosis and the interest in tracing EEG changes to brain electromagnetic signals (BEMS), seeking the cause of autism and the brain regions of its origin. The time- and frequency domain and principal component analysis (PCA) of these EEG signals with a Multilayer Perception Neural Network (MLP) identifies an autistic subject and helps improve classification accuracy. We show differences between a working brain and a relaxed brain, especially in the Alpha waves used for diagnosis. © 2012 IEEE.}, note = {cited By 11}, keywords = {Biomedical Engineering, Brain, Brain Regions, Classification Accuracy, Diseases, EEG Signals, Electromagnetic Signals, Electromagnetics, Electromagnetism, Frequency Domains, International Group, Multilayer Perception Neural Networks, Neuroimaging, Principal Component Analysis}, pubstate = {published}, tppubtype = {conference} } There has been a significant increase in the incidence of autism. We report the work on autism by our international group, on the growing attention paid to EEG based diagnosis and the interest in tracing EEG changes to brain electromagnetic signals (BEMS), seeking the cause of autism and the brain regions of its origin. The time- and frequency domain and principal component analysis (PCA) of these EEG signals with a Multilayer Perception Neural Network (MLP) identifies an autistic subject and helps improve classification accuracy. We show differences between a working brain and a relaxed brain, especially in the Alpha waves used for diagnosis. © 2012 IEEE. |
Testingadminnaacuitm2020-05-28T06:49:14+00:00
2019 |
Classification of adults with autism spectrum disorder using deep neural network Conference Institute of Electrical and Electronics Engineers Inc., 2019, ISBN: 9781728130415, (cited By 0). |
2013 |
Classification of emotional states from electrocardiogram signals: A non-linear approach based on hurst Journal Article BioMedical Engineering Online, 12 (1), 2013, ISSN: 1475925X, (cited By 42). |
2012 |
Autism, EEG and brain electromagnetics research Conference 2012, ISBN: 9781467316668, (cited By 11). |