SOURCE RECORD / S0197
Continuous tracking using deep learning-based decoding for noninvasive brain-computer interface
Continuous tracking using deep learning-based decoding for noninvasive brain-computer interface
- Authors or organization
- Forenzo D, Zhu H, Shanahan J, Lim J, He B
- Published date
- 2024-04-30
- Source type
- peer_reviewed_human_study
- Technology route
- non-invasive EEG motor imagery with deep-learning decoding
- Function or setting
- continuous 2D cursor control
- Rights status
- mirror_allowed / CC BY 4.0
Legacy aliases
—
Related topics
- BCI 最优方案
- 非侵入式 BCI
Evidence summary
Evidence: Online continuous pursuit with deep-learning decoders. Key figures: last-session NMSE: EEGNet and PointNet each versus traditional decoder p<0.0001; EEGNet versus PointNet p=0.2768; performance improved as subject-specific data accumulated. Limitations: Healthy experienced users; task-specific 2D virtual cursor; models accumulated session data; pretraining on other subjects did not significantly improve early performance; not patient or home evidence.
Key figures
last-session NMSE: EEGNet and PointNet each versus traditional decoder p<0.0001; EEGNet versus PointNet p=0.2768; performance improved as subject-specific data accumulated
Limitations
Healthy experienced users; task-specific 2D virtual cursor; models accumulated session data; pretraining on other subjects did not significantly improve early performance; not patient or home evidence
This file is a rights-reviewed, hash-verified mirror. This site is not the original publisher. File type：application/xml；File size：127384 bytes；SHA-256：0b661c9dcf8cd26cde9643498c1fb065082bf7b449795aa2d86fddb262b99b98；Verified at：2026-07-23T13:00:00+00:00。