Probabilistic Modelling of Sleep Stage and Apneaic Events in the University College of Dublin Database (UCDDB)
Bibliographic record
Abstract
Publicly available University College of Dublin Database (UCDDB) polysomnogram (PSG) patient cohort is analyzed and classified for archetypal sleep fragmentation in the presence of obstructive sleep apnea (OSA), the most severe sleep disorder. For comparison with Bianchi et al's analysis of the Sleep Heart Health Study (SHHS) polysomnogram cohort, we first analyze sleep stage hypnograms statically by considering Wake After Sleep Onset (WASO), Rapid Eye Movement (REM), and grouped non-REM stages (Stages 1, 2, and 3/4) probabilistic distributions. We test the universality of Bianchi et al's multi-exponential and power law models before proposing a third model with greater physical significance: Gaussian. In the absence of a control cohort without OSA, we separate the full PSG hypnogram into apneaic and `normal' components. From this analysis, we observe the rapid decay of sleep stage durations as a result of apneaic events and the fragmentation of the natural 1.5 to 2 hour sleep cycle. To refine our model of the distinctive apnea `fingerprint' on sleep stages, we consider a dynamic, one step Markov Chain model for WASO, REM, and separated NREM sleep stages (Stage 1, 2, 3, and 4) as this approach gives a better measure of sleep fragmentation. Our findings demonstrate that the presence of OSA alters both static and dynamic sleep characteristics which can feed into automatic detection of sleep stage identification and apnea event detection. Significantly, the probability of remaining in REM sleep is reduced from 81.0 percent during `normal' sleep to 13.7 percent following an apneaic event. We conclude by outlining immediate next steps for improving the apnea detection model for integration with our at-home automatic sleep monitoring device. Such a device will augment data collected from oversubscribed sleep clinics to address the undersampling of sleep disorder patient monitoring.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| 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) | 0.003 | 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".