Table 2.

Comparison of Proposed Model with Recent Similar Models.

S.No	Model	Methodology	Findings
1	ASD-DiagNet	Feature Map obtained from VAE [rs-fMRI]	Correlation coefficients can be influenced by head motion, the signal-to-noise ratio and individual differences in brain anatomy
2	DBN	Feature Map obtained from the Restricted Boltzmann Machine [rs-fMRI]	Correlation coefficients can be influenced by head motion, the signal-to-noise ratio and individual differences in brain anatomy
3	MMDC	Classification results of different modalities (phonation: voicing of vowels, voiced and unvoiced) are ensembled by blending or voting	Ensembling multiple modalities before classification can improve the accuracy compared to the ensembling of classification results
4	s-GCN	Nodes-> Atlas [rs-fMRI] Edges-> Phenotype data	The use of phenotype data for constructing edges might lead to a lack of heterogeneity since phenotype data do not consider variation in topographical locations
5	EigenGCN	Nodes-> Functional Connectivity Matrix (FCM) [rs-fMRI] Edges-> Graph Kernel Function [FCM]	Loss of crucial information about features due to multiple pooling layers
6	MVS-GCN	Brain graph constructed with multiple views from the subnetwork	The heterogeneity of the brain is not considered when forming subnetworks, which may result in a lack of functional connectivity between regions across subnetworks.
7	Proposed Model	Nodes-> Feature Maps obtained from Combined BS + Dual Regression Edges->Similarity Measure calculated using Radiomics from sMRI	-