Automated Sleep Spindle Analysis in Epilepsy EEG Using Deep Learning
Bibliographic record
Abstract
Sleep spindles, together with K-complexes, are hallmark oscillatory events observed in electroencephalographic (EEG) recordings during stage N2 sleep. Alterations in spindle characteristics, including frequency, amplitude, duration, and density, are frequently reported in epilepsy and may reflect underlying disturbances in thalamocortical network function. Quantitative analysis of these alterations has the potential to improve our understanding of epileptiform activity and support the development of clinically useful biomarkers. In this work, we present an automated framework for sleep spindle analysis in EEG recordings from both healthy subjects and patients with epilepsy. The framework integrates deep learning architectures (1D U-Net, SlumberNet, and SEED) with statistical evaluation methods to address two complementary tasks: (i) spindle segmentation and (ii) direct regression-based prediction of spindle characteristics. The proposed approach was validated on two datasets: the open-access Montreal Archive of Sleep Studies (MASS) and a custom clinical database of pediatric epilepsy patients acquired at the Video-EEG Laboratory “Genomed” (Moscow, Russia). Our results demonstrate that while both convolutional and hybrid recurrent–convolutional architectures achieve comparable overall F1-scores, their precision–recall profiles differ substantially. This enables a principled, context-specific selection of models, with U-Net favoring high sensitivity and SEED favoring high precision. Moreover, we show that segmentation-based pipelines consistently outperform direct regression (segmentation-free) approaches for characteristic prediction. These findings provide methodological guidance for the optimal deployment of deep learning models in sleep spindle analysis and establish a foundation for robust, automated, and clinician-independent EEG biomarkers in epilepsy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".