100 results for "signal classification" · page 8 of 10 · ranked by relevance
…p300 ## Signal Processing - **Feature extraction**: P300_ERP_detection ## Cross-Validation - **Method**: calibration…
This dataset comprises electroencephalographic recordings from 43 healthy subjects performing a visual P300-based brain-computer interface task using the Brain Invaders videogame paradigm. The experiment took place at GIPSA-lab, Grenoble, France, in 2015. Participants responded to an oddball stimulus presentation (36-symbol grid with 1 target, 35 non-targets) across three sessions with varying flash durations (50ms, 80ms, 110ms). EEG data were acquired using 32 wet electrodes at 512 Hz and processed with a calibration-free adaptive Riemannian classifier for real-time feedback. Resting state recordings (eyes closed, approximately one minute) were acquired before and after each session.
…computer interface applications, achieving mean classification accuracy of 73.3±8.9…
…p300 ## Signal Processing - **Feature extraction**: P300_ERP_detection ## Cross-Validation - **Method**: calibration…
This dataset comprises electroencephalographic recordings from 38 healthy subjects engaged in a multi-user P300-based Brain-Computer Interface experiment called Brain Invaders. Participants played in pairs across three conditions (Solo1, Solo2, and Collaboration) using a visual oddball paradigm with 36 symbols to elicit P300 responses. The study investigates inter-brain synchrony during collaborative BCI gaming, with 32-channel EEG recordings sampled at 512 Hz using wet electrodes in a standard 10-10 montage.
…p300 ## Signal Processing - **Feature extraction**: P300_ERP_detection ## Cross-Validation - **Method**: calibration…
This dataset comprises EEG recordings from a P300 visual matrix speller study comparing three unsupervised learning methods (Expectation-Maximization, Learning from Label Proportions, and their combination MIX) for brain-computer interface decoding. Twelve healthy participants performed a copy-spelling task using a modified 6×6 character grid extended with 10 # symbols as visual blanks (46 total symbols), recorded at 1000 Hz from 31 EEG channels. The study demonstrates that unsupervised learning methods can achieve performance comparable to supervised approaches without requiring calibration data.
…categories through single-trial EEG classification and representational similarity analysis. ## Dataset Summary…
A high-gamma EEG dataset comprising 14 healthy subjects performing motor imagery tasks (left hand, right hand, feet, and rest) recorded at 500 Hz with 128 channels. This is a BIDS-formatted derivative of the original dataset described in Schirrmeister et al. 2017, which was used to develop and validate deep convolutional neural networks for end-to-end EEG decoding. The derivative demonstrates that deep learning approaches can match or exceed traditional feature-based methods (FBCSP) while learning interpretable spectral power modulations in alpha, beta, and high-gamma frequency bands.
…p300 ## Signal Processing - **Feature extraction**: P300_ERP_detection ## Cross-Validation - **Method**: calibration…