Для улучшения точности ИМК предлагается раздельно моделировать внешний и внутренний "шум":
"External noises are caused by subjects’ movement or sensors’ instability, and internal noises result from the subjects’ random mental activities ... We introduce a plateau threshold to remove external noise and an unknown threshold set to detect unknown tasks to remove internal noise. Both unsupervised (such as K-Means) and supervised (Such as Random Forests, CNN, and RNN) learning algorithms are implemented in this HITL approach. We use the Thinking1 BCI experiments dataset with sixty subjects (available to academic researchers by request)."
Qu, X., Hickey, T.J. (2022). EEG4Home: A Human-In-The-Loop Machine Learning Model for EEG-Based BCI. In: Schmorrow, D.D., Fidopiastis, C.M. (eds) Augmented Cognition. HCII 2022. LNCS (LNAI), vol 13310. Springer, Cham. https://doi.org/10.1007/978-3-031-05457-0_14
"External noises are caused by subjects’ movement or sensors’ instability, and internal noises result from the subjects’ random mental activities ... We introduce a plateau threshold to remove external noise and an unknown threshold set to detect unknown tasks to remove internal noise. Both unsupervised (such as K-Means) and supervised (Such as Random Forests, CNN, and RNN) learning algorithms are implemented in this HITL approach. We use the Thinking1 BCI experiments dataset with sixty subjects (available to academic researchers by request)."
Qu, X., Hickey, T.J. (2022). EEG4Home: A Human-In-The-Loop Machine Learning Model for EEG-Based BCI. In: Schmorrow, D.D., Fidopiastis, C.M. (eds) Augmented Cognition. HCII 2022. LNCS (LNAI), vol 13310. Springer, Cham. https://doi.org/10.1007/978-3-031-05457-0_14