Reproduce “CANDLE: Cortical Null-Space Decomposition for Noninvasive Brain Source Imaging” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shuntaro Suzuki, Yuiga Wada, Komei Sugiura
CANDLE uses subject-specific MRI-derived lead-field matrices to recover the source component constrained by scalp EEG and learn only the unobservable component. In simulated data, it achieved an HD95 of 37.3 mm and Dice score of 0.48, improving on the second-best baseline. In intracranial stimulation localization, it had the lowest overall mean peak-distance error, but not significantly better than GBF or sLORETA.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.