MHATC: Autism Spectrum Disorder identification utilizing multi-head attention encoder along with temporal consolidation modules

TL;DR

This paper introduces MHATC, a novel deep learning model using multi-head attention and temporal modules to improve ASD classification from resting-state fMRI data, addressing limitations of existing methods.

Contribution

The paper presents a new architecture combining multi-head attention and temporal consolidation for more accurate and efficient ASD diagnosis from fMRI data.

Findings

Robust and computationally efficient classification performance.

Addresses limitations of current deep neural network solutions.

Potential for broad adoption in clinical settings.

Abstract

Resting-state fMRI is commonly used for diagnosing Autism Spectrum Disorder (ASD) by using network-based functional connectivity. It has been shown that ASD is associated with brain regions and their inter-connections. However, discriminating based on connectivity patterns among imaging data of the control population and that of ASD patients' brains is a non-trivial task. In order to tackle said classification task, we propose a novel deep learning architecture (MHATC) consisting of multi-head attention and temporal consolidation modules for classifying an individual as a patient of ASD. The devised architecture results from an in-depth analysis of the limitations of current deep neural network solutions for similar applications. Our approach is not only robust but computationally efficient, which can allow its adoption in a variety of other research and clinical settings.