Machine learning approaches to predicting whether muscles can be elicited via TMS
Background: Transcranial magnetic stimulation (TMS) is a valuable technique for assessing the function of the motor cortex and cortico-muscular pathways. TMS activates the moto-neurons in the cortex, and this activation is transmitted through the cortico-muscular pathway, after which it can be measured as a motor evoked potential (MEP) in the muscles. The position and orientation of the TMS coil and the intensity used to deliver a TMS pulse are considered central TMS setup parameters influencing the presence/absence of MEPs. New Method: We sought to predict the presence of MEPs from TMS setup parameters using machine learning. We trained different machine learners using either within-subject or between-subject designs. Results: We obtained prediction accuracies of on average 77% and 65% with maxima up to up to 90% and 72% within and between subjects, respectively. Across the board, a bagging ensemble appeared to be the most suitable approach to predict the presence of MEPs, although a comparably simple logistic regression model also performed well. Conclusions: While the prediction between subjects clearly leaves room for improvement, the within-subject performance encourages to supplement TMS by machine learning to improve its diagnostic capacity with respect to motor impairment.