Imbalance-aware Machine Learning for Epileptic Seizure Detection
Bibliographic record
Abstract
Automatic epileptic seizure detection is a challenging task that could cope with sudden seizures and help epileptic patients to have a normal life. The electroencephalography (EEG) recording remains the most common method used for detecting epileptic seizures. The precision and accuracy of seizure detection are the most important elements in automatic EEG-based seizure detection systems, which could be achieved by training the classification models with relevant features. In this work, we propose a robust machine learning framework for epileptic seizure detection from EEG data. Imbalance class problem and high dimensional feature space issue have been handled for classification. Our approach has been tested on the largest EEG database (The Temple University Hospital EEG Seizure Corpus, TUSZ). A comparative study on three categories of data balancing techniques: costsensitive learning (weighting), oversampling and under sampling has been made. An efficient feature selection algorithm based on feature interaction graph analysis has been used for selecting minimal number of relevant inputs before classification. Results in terms of accuracy and area under the curve (AUC), have showed that the features subset selected using the graph-based method, balanced by the Synthetic Minority Over Sampling method (SMOTE) achieved the highest classification performance using random forest classifier.
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 | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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".