Sohrabpour Abbas, He Bin
Department of Biomedical Engineering, Carnegie Mellon University, United States.
Curr Opin Biomed Eng. 2021 Jun;18. doi: 10.1016/j.cobme.2021.100277. Epub 2021 Mar 1.
Electrophysiological source imaging (ESI) has been successfully employed in many brain imaging applications during the last 20 years. ESI estimates of underlying brain networks provide millisecond resolution of dynamic brain processes; yet, it remains to be a challenge to further improve the spatial resolution of ESI modality, in particular on its capability of imaging the extent of underlying brain sources. In this review, we discuss the recent developments in signal processing and machine learning that have made it possible to image the extent, i.e. size, of underlying brain sources noninvasively, using scalp electromagnetic measurements from electroencephalogram (EEG) and magnetoencephalogram (MEG) recordings.
在过去20年中，电生理源成像（ESI）已成功应用于许多脑成像应用中。对潜在脑网络的ESI估计可提供动态脑过程的毫秒级分辨率；然而，进一步提高ESI模态的空间分辨率，特别是其对潜在脑源范围进行成像的能力，仍然是一项挑战。在这篇综述中，我们讨论了信号处理和机器学习方面的最新进展，这些进展使得利用脑电图（EEG）和脑磁图（MEG）记录的头皮电磁测量结果，以非侵入性方式对潜在脑源的范围（即大小）进行成像成为可能。
