<ul><li>AbsoluteNet is a new deep learning neural network designed to classify auditory event-related responses recorded using functional near-infrared spectroscopy.</li><li>The network utilizes spatio-temporal convolution and customized activation functions to achieve superior performance.</li><li>AbsoluteNet outperformed existing models like fNIRSNET, MDNN, DeepConvNet, and ShallowConvNet, achieving 87.0% accuracy, 84.8% sensitivity, and 89.2% specificity in binary classification.</li><li>The study highlights the effectiveness of AbsoluteNet in decoding hemodynamic responses related to auditory processing, emphasizing the significance of spatio-temporal feature aggregation and customized activation functions for fNIRS dynamics.</li></ul>

AbsoluteNet: A Deep Learning Neural Network to Classify Cerebral Hemodynamic Responses of Auditory Processing

Discover more