Predicting Brain Electrical Stimulation Outcome in Stroke by Clinical-inspired Hybrid Graph Convolutional Autoencoder

Noninvasive brain stimulation (NIBS) has gained lots of attention from both academics and clinical usage. Its curative effect shows positive feedback in different kinds of neurological and ophthalmological disorders. Stroke is one of them that could benefit from this new technology. However, the unknown underlying mechanism of brain stimulation hinders our further exploration of brain recovery. This study proposes a hybrid graph convolutional autoencoder (HGCAE) to predict stroke recovery potential after electrical stimulation therapy. The results show that using HGCAE based on brain network measures achieved an overall sensitivity of 91% of predicting recovery following NIBS intervention. This result may help predict the potential outcome of brain modulation in stroke patients and allows us gain more insight into clinical interventions.