Brain Decoding over the MEG Signals Using Riemannian Approach and Machine Learning

Brain decoding is an emerging approach for understanding the face perception mechanism in human brain. Face visual stimuli and are considered as a challenging ongoing research of neuroscience field. In this study, face/scrambled stimulations were implemented over sixteen participants to be decoded or scrambled classification using machine learning (ML) algorithms via magnetoencephalography (MEG) signals. This noninvasive high spatial/temporal resolution signal neurophysiological technique which measures magnetic fields generated by neuronal activity The Riemannian was used highly promising feature extraction technique. Then Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Convolutional Neural Network (CNN) employed deep algorithms, Linear Discriminant Analysis (LDA) Quadratic (QDA) shallow algorithms. improved performances very encouraging, especially LSTM GRU have achieved 92.99% 91.66% accuracy 0.977 0.973 area under curve (AUC) scores, respectively. Moreover, CNN has yielded 90.62% accuracy. As our best knowledge, outcomes usage on MEG dataset signals from 16 critical expand literature brain after stimuli. And study first attempt with these methods systematic comparison. MEG-based Brain-Computer Interface (BCI) approaches may also Internet Things (IoT) applications, including biometric authentication, thanks specific individual’s brainwaves.