Artificial Intelligence and Communication Technologies
TransLearning ASD: Detection of Autism Spectrum Disorder Using Domain Adaptation and Transfer Learning-Based Approach on RS-FMRI Data
Authors: Samreen Singh, Deepti Malhotra and Mehak Mengi
Publishing Date: 03-04-2023
ISBN: 978-81-955020-5-9
Abstract
Autism Spectrum Disorder abbreviated as ASD, is a complex neuro-developmental disease specifically linked to nervous system that influences patients’ communicationand social behavior. Traditional clinical techniques used for the discovery of ASD fall short of definite and early ASD diagnosis. Consequently, biomarkers have been introduced in the field of ASD diagnosis and particularly, resting-state functional Magnetic Resonance Imaging (rs-fMRI) has posed as a valuable biomarker. Researchers have focused on utilizing the vast span of Artificial Intelligence techniques in combination with rs-fMRI, to build an effective framework for ASD detection. However, these systems have not been able to generalize to a larger set of patients, because of theheterogeneity in the available f-MRI dataset for ASD. Motivated from the aforementioned discussion, this study performs a comprehensive literature review of the existing systems covering a period of 2019-2021, thereby identifying several research gaps. To overcome the effect of existing implications, this paper expounds a TransLearning ASD framework which will achieve normalization of the heterogeneous fMRI data using domain adaptation followed by transfer learning technique for effective ASD prediction and to overcome the model generalization problem.
Keywords
Machine Learning, Deep learning, ABIDE, ASD, Functional MRI, Domain Adaptation, Transfer Learning etc.
Cite as
Samreen Singh, Deepti Malhotra and Mehak Mengi, "TransLearning ASD: Detection of Autism Spectrum Disorder Using Domain Adaptation and Transfer Learning-Based Approach on RS-FMRI Data", In: Saroj Hiranwal and Garima Mathur (eds), Artificial Intelligence and Communication Technologies, SCRS, India, 2023, pp. 863-871. https://doi.org/10.52458/978-81-955020-5-9-81