Datensatz

Zusammenfassung

Background: Transcranial magnetic stimulation (TMS) modulates brain activity by inducing electric fields (E-fields) that can elicit action potentials in cortical neurons. Neuronal responses to TMS depend not only on the magnitude of the induced E-field but also on various physiological factors. In this study, we incorporated a novel average response model that efficiently estimates the firing threshold of neurons based on their orientation relative to the applied E-field, thereby advancing TMS mapping for motor function. Methods: We conducted a regression-based TMS mapping experiment with fourteen subjects to localize cortical origins of motor evoked potential (MEP) on the first dorsal interosseous (FDI) muscle. Firing thresholds were estimated for excitatory neurons in cortical layers 2/3 and 5 via an average response model. Regression was performed between MEPs and three E-field quantities: the magnitude (magnitude model), the normal component (cosine model), and the effective E-field, which scales the E-field magnitude based on the firing thresholds specific to the neuronal orientation (neuron model). To validate, we applied TMS to ten subjects with optimized coil placements based on these three models to determine which model could yield the highest MEPs. Results: The magnitude and neuron models performed similarly, while the cosine model showed lower explained variance in regression results, required more TMS trials for stable mapping, and yielded the lowest MEP in the validation. Conclusion: This study is the first to advance TMS modeling by incorporating neuron-specific factors at the individual level. Results show that on the motor cortex, the magnitude model is, as expected, a good approximation of cortical TMS effects as it shows similar results as the neuron model. In contrast, the classic cosine model exhibited lower performance and required more TMS trials for stable results, and is not recommended for future studies.