Sleep Apnea Detection by Tracheal Motion and Sound, and Oximetry via Application of Deep Neural Networks
Bibliographic record
Abstract
Purpose: Sleep apnea (SA) is highly prevalent, but under diagnosed due to inaccessibility of sleep testing. To address this issue, portable devices for home sleep testing have been developed to provide convenience with reasonable accuracy in diagnosing SA. The objective of this study was to test the validity a novel portable sleep apnea testing device, BresoDX1, in SA diagnosis, via recording of trachea-sternal motion, tracheal sound and oximetry. Patients and Methods: Adults with a suspected sleep disorder were recruited to undergo in-laboratory polysomnography (PSG) and a simultaneous BresoDX1 recording. Data from BresoDX1 were collected and features related to breathing sounds, body motions and oximetry were extracted. A deep neural network (DNN) model was trained with 61-second epochs of the extracted features to detect apneas and hypopneas from which an apnea-hypopnea index (AHI) was calculated. The AHI estimated by BresoDX1 (AHI breso ) was compared to the AHI determined from PSG (AHI PSG ) and the sensitivity and specificity of SA diagnosis were assessed at an AHI PSG ≥ 15. Results: Two-hundred thirty-three participants (mean ± SD) 50 ± 16 years of age, with BMI of 29.8 ± 6.6 and AHI of 19.5 ± 22.7, were included. There was a strong relationship between AHI breso and AHI PSG (r = 0.91, p < 0.001). SA detection for an AHI PSG ≥ 15 was highly sensitive (90.0%) and specific (85.9%). Conclusion: We conclude that the DNN model we developed via recording and analyses of trachea-sternal motion and sound along with oximetry provides an accurate estimate of the AHI PSG with high sensitivity and specificity. Therefore, BresoDX1 is a simple, convenient and accurate portable SA monitoring device that could be employed for home SA testing in the future. Keywords: sleep apnea, portable sleep testing, tracheal acoustics
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) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 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 | 0.001 |
| 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 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".