An enhanced analysis of non-invasive human brain stimulation prediction for stroke patients using closed loop control scheme

This paper presents a novel closed loop control scheme for non-invasive human brain stimulation (NIBS) prediction in stroke patients. The proposed scheme utilizes an artificial neural network (ANN) model based on parameters derived from electroencephalography recordings of stroke patients’ motor cortex activity. The ANN model was trained to predict the future motor cortex activation based on the past patient’s brain activity. Then, the model was tested on a simulated clinical setting to assess its accuracy and efficacy. The results showed that the model was superior to existing methods in terms of accuracy and efficacy. Furthermore, the proposed scheme was able to dynamically adjust the parameters of NIBS stimulation based on the predictions made by the ANN model. This study provides a proof of concept for using closed loop control schemes to predict and optimize NIBS stimulation in stroke patients.