Skip to main content
. Author manuscript; available in PMC: 2023 Dec 1.
Published in final edited form as: Neuroimage. 2022 Oct 21;264:119705. doi: 10.1016/j.neuroimage.2022.119705

Fig.1.

Fig.1.

A self-supervised deep learning model for computing TMS E-field. A deep neural network is applied to learn a mapping from individual head model and TMS induced primary E-field to the TMS electric scalar potential, and the network is optimized by a loss function determined by an energy function (Wang and Eisenberg, 1994).