Automatic Sleep Spindle Detection Using SMOTE and Composite Features with SWT and Adaboost
Bibliographic record
Abstract
Sleep Spindles contribute to diagnosing several brain-related diseases like sleep apnea, major depression, etc.Hence, sleep spindle detection from Electroencephalogram (EEG) has gained significant research interest in the bio-medical signaling field.Existing methods use template matching and machine learning algorithms for spindle detection.In the template matching methods, the additional and continuous tuning of the threshold creates an unnecessary computational burden.In the machine learning-based method, significant problems such as data imbalance and less discrimination due to fewer and inappropriate features are addressed.To address these issues, this research presents a novel automatic spindle recognition approach that uses the Synthetic Minority Oversampling Technique (SMOTE) to balance the dataset and integrates time-domain and frequency-domain information for effective feature extraction.The Synchrosqueezed Wavelet Transform (SWT) is utilized for accurate frequency domain feature extraction, while the Adaboost(Adaptive Boosting)algorithm is implemented for classification.This method, evaluated using the publicly accessible Montreal Archives of Sleep Studies Cohort 1 (MASS-C1) dataset, significantly outperforms existing methods like as SpindleU-Net, Convolutional MIL (Multiple Instance Learning), SST-RUSBoost (Synchrosqueezed Transform -Random Under-Sampling Boosting), and MuFF-E(Multi-Feature Fusion and Ensemble), with an F-score of 75%, Sensitivity of 78%, and Positive Predictive Value of 73%.The findings illustrate the superiority of the proposed method in addressing data imbalance and improving detection accuracy.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".