A semi-supervised deep learning algorithm for abnormal EEG identification
This addresses the challenge of automating EEG analysis to aid neurologists, though it appears incremental as it builds on semi-supervised methods.
The paper tackles the problem of needing large labeled datasets for training EEG analysis systems by proposing a semi-supervised learning workflow that can make predictions with as few as 5 labeled examples, reducing the workload for neurologists.
Systems that can automatically analyze EEG signals can aid neurologists by reducing heavy workload and delays. However, such systems need to be first trained using a labeled dataset. While large corpuses of EEG data exist, a fraction of them are labeled. Hand-labeling data increases workload for the very neurologists we try to aid. This paper proposes a semi-supervised learning workflow that can not only extract meaningful information from large unlabeled EEG datasets but also make predictions with minimal supervision, using labeled datasets as small as 5 examples.