Work

Geometric Features for Brain-Computer Interface Classification: Benchmark Results on PhysioNet Motor Imagery

Year: 2026

Type: article

Abstract: OmniSciences Technical Report TR-2026-003 We present benchmark results for proprietary geometric feature extraction applied to EEG-based brain-computer interface (BCI) motor imagery classification. The method operates on spatial covariance matrices on the SPD manifold. Key results: PhysioNet (109 subjects, 64 channels): +0.78pp over pyRiemann baseline (p=0.0048, Wilcoxon signed-rank test) Strongest advantage at low channel counts (n=3-8), relevant for consumer BCI Cross-subject transfer: +8.5pp over standard MDM Patent-pending (U.S. Provisional Application No. 64/004,047).

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.19192337

Author: OmniSciences Research

Institution: Omni Sciences (United States)

Language: English

Topic: EEG and Brain-Computer Interfaces

Subfield: Cognitive Neuroscience

Field: Neuroscience

Domain: Life Sciences

Open Access status: green

OpenAlex ID: W7140088412

FWCI:

Cites:

Cited by:

Related to: