Automated and noninvasive in vivo mapping of brain functions in the human cortex

Mapping forms the basis for understanding the functional organization of the brain and for developing targeted interventions in neurological and psychiatric disorders. Transcranial magnetic stimulation (TMS) is a non-invasive tool that allows causal mapping of cortical functions involved in motor control, speech, and connectivity. Since manual mapping procedures limit standardization, we present an automated cortical mapping method that integrates electronically controlled multi-locus TMS with a Bayesian optimization algorithm for identifying stimulation targets efficiently and objectively through a closed-loop framework. We localized individual motor representation area peaks with high accuracy (average 2.4 mm and 5° deviation from the optimum) in only approximately 4 minutes. This represents a substantial advancement over current mapping approaches, providing the means to greatly accelerate and standardize both research and clinical protocols while maintaining high reliability. We anticipate that our approach can further extend to map other cortical functions that cannot be easily analyzed online by a human operator.