MétaCan
Menu
← Back to cohort

Distinct Biometric Differences in Obstructive Sleep Apnea Patients Phenotyped Based on Airflow Patterns Highlight Potential for Customized Treatment Algorithm Development

2025· article· en· W4410275688 on OpenAlexaff
Hamed Hanafi, Michelle J. Hickey, Kazi Antor Hasan, S. Driscol, Stephen Lowe, Ingo Fietze, Thomas Penzel, Reena Mehra, Sanja Jelić, Debra Morrison

Bibliographic record

VenueAmerican Journal of Respiratory and Critical Care Medicine · 2025
Typearticle
Languageen
FieldMedicine
TopicObstructive Sleep Apnea Research
Canadian institutionsQueen Elizabeth II Health Sciences CentreDalhousie University
Fundersnot available
KeywordsMedicineObstructive sleep apneaAirflowBiometricsSleep apneaAlgorithmIntensive care medicineArtificial intelligenceInternal medicineComputer science

Abstract

fetched live from OpenAlex

Abstract Rationale: Obstructive sleep apnea (OSA) affects approximately 900 million people worldwide. While positive airway pressure (PAP) therapy is effective in treating OSA, one-year adherence remains below 50%. It has been suggested that patient discomfort caused by high air pressure settings may be a key obstacle to patient adherence. Our innovation, continuous management of airway pressure (cMAP®), utilizes artificial intelligence (AI) to predict and prevent OSA events to reduce therapy pressure and enhance patient comfort while maintaining treatment efficacy1. It may be possible to develop versions of cMAP® which have been tuned to specific phenotypes of patients with OSA on PAP therapy. Therefore, it was our objective to investigate if specific airflow-based patient phenotypes existed in patients diagnosed with OSA. Methods: In a previous study, we completed a randomized crossover study that included 45 patients (Age: 52±9 years, BMI: 41±12 kg/m2, Gender: 27 Male – 18 Female) diagnosed with mild to severe OSA2. In an attempt to discover airflow-based patient phenotypes, we employed sparse projection pursuit3, an unsupervised exploratory data analysis method, to reveal distinct clusters of patients. Additionally, we used logistic regression in conjunction with recursive feature elimination to determine which patient biometric and treatment variables measured by the PAP machine or SleepImage RingTM were most predictive of each potential patient phenotype. Results: We identified three well-separated clusters associated with patient PAP airflow patterns (Figure 1). Biometric and treatment variables which were shown to be most predictive of patient phenotype were obstructive apnea index, central apnea index, tidal volume, diagnosed apnea-hypopnea index (AHI), flow limitation, respiratory disturbance index (RDI), fragmentation, and sleep quality index (SQI). Type I was characterized by lower AHI/RDI while Type II was characterized by higher SQI and tidal volume with lower fragmentation. Type III was characterized by higher AHI/RDI and fragmentation with lower SQI and tidal volume. Conclusions: Our preliminary results demonstrate that there may exist specific airflow-based phenotypes of patients with OSA. We believe this represents the first step towards the development of PAP therapy directly tailored to the specific airflow patterns and treatment needs of particular groups of patients. Overall, the development of these tailored treatment regimes may further augment the comfort and health benefits attributable to cMAP® therapy. 1. Hanafi et al. ERS. 2024;64:68. 2. Hanafi et al. Am J Respir Crit Care Med. 2024;209:A4570. 3. Driscoll et al. Anal. Chem. 2020;92,1755−1762

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.004
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.002
Threshold uncertainty score0.007

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.004
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.000
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0020.000

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.

Opus teacher head0.016
GPT teacher head0.311
Teacher spread0.295 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueAmerican Journal of Respiratory and Critical Care Medicine→Same topicObstructive Sleep Apnea Research→French-language works237,207→