Detection of focal impaired awareness seizures using a biometric shirt
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
OBJECTIVE: In recent years, seizure detection using wearable technology has gained significant attention in research. Most studies, however, have focused on detecting generalized or focal to bilateral tonic-clonic seizures. This study evaluates the feasibility of using a biometric shirt to detect focal impaired awareness seizures (FIAS) by monitoring respiratory, electrocardiography, and accelerometry signals. METHODS: Patients with epilepsy were recruited at the University of Montreal Hospital Center epilepsy monitoring unit. Seizures were annotated by epileptologists based on simultaneous video-electroencephalographic recordings, blinded to the shirt data. Features were extracted from the respiratory, accelerometry, and electrocardiography signals using varying window sizes and steps. An XGBoost classifier was trained and tested using a nested leave-one-subject-out cross-validation. Post-processing included a firing power regularization method to reduce false alarms. RESULTS: We recorded 113 FIAS from 27 patients who wore the shirt continuously for over 4750 hours. Using a firing power threshold of 0.65, we detected 71 seizures, resulting in a mean sensitivity of 66%, a 15% time in warning, and a false alarm rate (FAR) of 30 per 24 hours. A firing power threshold of 0.85 allowed us to reduce false alarms (8% time in warning, FAR of 21 per 24 hours) but resulted in a lower sensitivity of 57%. Performances varied across patients: sensitivity was high and FAR was low for some patients and vice versa for others, indicating variability in algorithm effectiveness across patients. SIGNIFICANCE: Our results demonstrate that detecting FIAS with a connected shirt could be feasible for certain patients, although the rate of false alarms remains an issue. Designing a personalized algorithm and selecting patients who exhibit significant physiological changes during seizures could make wearable-based FIAS detection more viable in the near future. PLAIN LANGUAGE SUMMARY: Novel mobile health technologies could transform epilepsy care by enabling continuous monitoring of seizures in everyday life. In this study, we used a biometric shirt to detect seizures with impaired awareness automatically. We used the shirt to measure breathing, heart activity, and movement in 27 patients with epilepsy in a hospital setting. Our algorithm detected up to two-thirds of seizures correctly. However, the number of incorrect alarms remains relatively high, with variable performances between patients. While the technology showed potential, these challenges highlight the need for further improvements and personalized care plans.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it