sleep-edf

CCRRSleepNet: A Hybrid Relational Inductive Biases Network for Automatic Sleep Stage Classification on Raw Single-Channel EEG

EEG & Hypnogram Sleep Quality Scoring with CNN + LSTM

End-to-end sleep-stage classification using EEG, EOG, and EMG signals with a sub-100K parameter model designed for edge deployment.

classification and dim reduction methods with traditional machine learning techniques

Clasificación de estadios de sueño usando EEG, EOG y EMG del dataset Sleep-EDF Expanded // Sleep stage classification using EEG, EOG, and EMG from the Sleep-EDF Expanded dataset

Traceable machine learning on physiological and neural signals: wrist, chest and sleep-EEG pipelines, evaluated subject-wise, with provenance from every prediction back to the raw window.

Closed-loop BCI pipeline for automated lucid dream induction via real-time EEG sleep stage classification | Part of LUCID: Reality?

Exploratory sleep staging and fragmentation analysis from Sleep-EDF PSG data

Sleep stage classification from raw EEG/EOG using a spatial-temporal CNN (Chambon 2018 variant). Trained on PhysioNet SleepEDF-78 with MNE-Python preprocessing, ICA artifact removal, and PyTorch. Achieves ~0.72 Cohen's Kappa on subject-wise held-out test set.

Interpretable EEG sleep-stage classification (Sleep-EDF) with subject-wise cross-validation and HMM temporal smoothing.

Sleep stage classification from Sleep-EDF EEG signals using a leakage-safe, subject-grouped nested cross-validation pipeline with MLflow-tracked reproducibility.

TempoSleep is a context-aware framework for automatic single-channel EEG sleep staging. It combines multi-scale temporal feature extraction with hierarchical temporal modeling to capture local and long-range dependencies, with particular emphasis on N1-stage recognition.

A controlled benchmark of ten hyperparameter optimization algorithms searching the same CNN–LSTM architecture on two EEG sleep datasets (Sleep-EDF, HMC) under a fixed budget. 1,992 evaluations, selected on validation loss and scored on macro F1, which rank the methods differently.

24/7 multimodal bio-sensing wearable platform: tiered acquisition (PPG/IMU continuous → EEG/fNIRS rest), edge AI triage, LSL/XDF sync, validated on WESAD/MIT-BIH/Sleep-EDF