Classification of Autism Spectrum Disorder Using Edge-Weight Enhanced Graph Attention Network With Multiple Features of Resting-State fNIRS Signals.

Researchers developed a new computer method to identify autism in children using a brain imaging technique called fNIRS. They studied brain activity patterns in 47 children (22 without autism, 25 with autism) while they were resting. The new method was very accurate at telling the difference between children with and without autism, correctly identifying autism 97.92% of the time.

This study developed a novel machine learning approach using Edge-Weight Enhanced Graph Attention Network (EWE-GAT) to identify autism spectrum disorder (ASD) in children using functional near-infrared spectroscopy (fNIRS) brain imaging. The research included 22 typically developing children and 25 children with ASD, measuring brain activity in bilateral temporal lobes during rest. Seven different features were analyzed, including blood oxygenation patterns and brain connectivity measures. The proposed method achieved exceptional classification performance with 97.92% accuracy, 100% sensitivity, 96.43% precision, and 98.08% F1 score, significantly outperforming traditional machine learning and neural network approaches.

fNIRS-based machine learning shows promise as an objective ASD identification tool. However, larger validation studies are needed before clinical implementation. Current findings support continued research into neuroimaging-based autism detection methods.

Small sample size (47 total participants). Single study without independent validation. Unclear generalizability across age groups, autism severity levels, or diverse populations. Method requires specialized neuroimaging equipment and technical expertise.

Functional near-infrared spectroscopy (fNIRS), as a noninvasive brain imaging modality, has shown great potential for autism spectrum disorder (ASD) identification combined with machine learning. In this work, we proposed an ASD identification method using edge-weight enhanced graph attention network (EWE-GAT) with multiple features in resting-state fNIRS signals measured from the bilateral temporal lobes on 22 typically developing (TD) children and 25 children with ASD. Seven features were selected for the EWE-GAT model, including five node features: the coupling between oxygenated hemoglobin (HbO) and deoxygenated hemoglobin (Hb) fluctuations, sample entropy for HbO and Hb, and average resting-state functional connectivity (RSFC) for HbO and Hb of each channel, and two edge features: RSFC between each channel pair for both HbO and Hb. With the proposed method, high accurate classification was achieved with 97.92% accuracy, 100% sensitivity, 96.43% precision, and 98.08% F1 score, outperforming usually used traditional machine learning and convolutional neural network models.