Transformer Model for Functional Near-Infrared Spectroscopy Classification

• Published in: IEEE journal of biomedical and health informatics Vol. 26; no. 6; pp. 2559 - 2569
• Main Authors: Wang, Zenghui, Zhang, Jun, Zhang, Xiaochu, Chen, Peng, Wang, Bing
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.06.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Artificial neural networks; Brain; Brain-computer interface (BCI); Brain-Computer Interfaces; classificat- ion; Classification; Classifiers; Computer applications; Computer vision; Convolution; Deep learning; end-to-end; Feature extraction; Functional near-infrared spectroscopy; functional near-infrared spectroscopy (fNIRS); Hemodynamics; Human-computer interface; Humans; I.R. radiation; Implants; Infrared spectra; Infrared spectroscopy; Long short-term memory; Machine learning; Medical imaging; Movement; Natural language processing; Near infrared radiation; Neural networks; Neural Networks, Computer; Neuroimaging; Performance enhancement; Preprocessing; preprocessing module; Representations; self-attention; Spectroscopy, Near-Infrared - methods; Spectrum analysis; Task analysis; transformer; Transformers

Functional near-infrared spectroscopy (fNIRS) is a promising neuroimaging technology. The fNIRS classification problem has always been the focus of the brain-computer interface (BCI). Inspired by the success of Transformer based on self-attention mechanism in the fields of natural language processin...