Application of Blind Separation of Sources to Optical Recording of Brain Activity

摘要: In the analysis of data recorded by optical imaging from intrinsic signals (measurement changes light reflectance cortical tissue) removal noise and artifacts such as blood vessel patterns is a serious problem. Often bandpass filtering used, but underlying assumption that spatial frequency exists, which separates mapping component other components (especially global signal), questionable. Here we propose alternative ways processing data, using blind source separation techniques based on decorrelation data. We first perform benchmarks artificial in order to select way processing, most robust with respect sensor noise. then apply it recordings experiments macaque primary visual cortex. show our BSS technique able extract ocular dominance orientation preference maps single condition stacks, for where standard post-processing procedures fail. Artifacts, especially patterns, can often be completely removed maps. summary, method extended superior recording