EEG-based Processing and Classification Methodologies for Autism Spectrum Disorder: A Review

Abstract

Autism Spectrum Disorder is a lifelong neurodevelopmental condition which affects social interaction, communication and behaviour of an individual. The symptoms are diverse with different levels of severity. Recent studies have revealed that early intervention is highly effective for improving the condition. However, current ASD diagnostic criteria are subjective which makes early diagnosis challenging, due to the unavailability of well-defined medical tests to diagnose ASD. Over the years, several objective measures utilizing abnormalities found in EEG signals and statistical analysis have been proposed. Machine learning based approaches provide more flexibility and have produced better results in ASD classification. This paper presents a survey of major EEG-based ASD classification approaches from 2010 to 2018, which adopt machine learning. The methodology is divided into four phases: EEG data collection, pre-processing, feature extraction and classification. This study explores different techniques and tools used for pre-processing, feature extraction and feature selection techniques, classification models and measures for evaluating the model. We analyze the strengths and weaknesses of the techniques and tools. Further, this study summarizes the ASD classification approaches and discusses the existing challenges, limitations and future directions.

Manifold Learning Analysis for Allele-Skewed DNA Modification SNPs for Psychiatric Disorders

“…For example, investigating the possible impacts of individual SNPs on SCZ subtype generation and classification can be a more practical approach to unveil SCZ pathological subdivisions along with new learning techniques such as LIME and ResNet for high-dimensional SNP data [74], [75]. In addition, we will consider the use of EEG and fMRI data to comprehensively explore the pathogenesis of mental illness along with bioinformatics data [76], [77].…”

Children ASD Evaluation Through Joint Analysis of EEG and Eye-Tracking Recordings With Graph Convolution Network

“…Multi-domain features are captured to highlight the characteristics of EEG signals, e.g., information entropy, timefrequency analysis, and synchronization analysis. Specifically, information entropy features are first obtained by measuring the complexity of the original EEG signal individually along each channel, e.g., approximate entropy (ApprEn, Meedeniya et al, 2019), sample entropy (SampEn, Liu et al, 2017), and permutation entropy (PermEn, Kang et al, 2019). Then, the original EEG is processed by short-time Fourier transforms (STFT) to capture the time-frequency representation, e.g., entropy feature.…”

“…The time information like such as total fixation duration on certain AOI is recorded to quantify each child's engagement for each AOI, where a 60 ms threshold is applied to avoid counting unconscious gazing. Note that random forest algorithm (Meedeniya et al, 2019) is utilized to filter out the most significant features.…”

Ultrasound Nerve Segmentation Using Deep Probabilistic Programming

“…The DPPL-based approach presented in this study can be extended to other domains as well. For instance, DPPL can be used in the diagnosis of psychophysiological disorders using neuroimaging data [27,28]. Figure 8 illustrates a possible abstract design of such a generalized neuroscience decision support system model that can identify different diseases and disorders using neuroimaging data.…”