Related Experiment Video
Updated: Aug 14, 2026
07:54
Drug-Induced Sleep Endoscopy (DISE) with Target Controlled Infusion (TCI) and Bispectral Analysis in Obstructive Sleep Apnea
Published on: December 6, 2016
Development of an Intelligent Clinical Decision Support System for Predicting One-Year CPAP Adherence in Patients
Emma López-Prado1, Manuel Casal-Guisande2,3,4, Mar Mosteiro-Añón1,3,5
1School of Industrial Engineering, University of Vigo, 36310 Vigo, Spain.
Journal of Clinical Medicine
|August 13, 2026
Summary
Machine learning can predict continuous positive airway pressure (CPAP) adherence in obstructive sleep apnea (OSA) patients. Including early adherence data significantly improves prediction accuracy, enabling personalized treatment strategies.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Medicine
- Sleep Medicine
Background:
- Obstructive sleep apnea (OSA) is a common chronic condition.
- Continuous positive airwayway pressure (CPAP) is a primary treatment for OSA.
- Patient adherence to CPAP therapy is crucial for treatment effectiveness.
Purpose of the Study:
- To develop and evaluate a machine learning (ML)-based clinical decision support system.
- To predict CPAP adherence after one year of treatment.
- To enable early identification of patients at risk for low adherence.
Main Methods:
- A cohort of 200 OSA patients was utilized, split into training (n=160) and testing (n=40) sets.
- Two prediction scenarios were defined: pre-treatment variables (Scenario A) and pre-treatment plus first-month adherence (Scenario B).
- Recursive feature elimination was used for variable selection, and Random Forest was the chosen classifier.
Main Results:
- Scenario A (pre-treatment variables) achieved an Area Under the Curve (AUC) of 0.71.
- Scenario B (including first-month adherence) significantly improved prediction with an AUC of 0.91.
- The Random Forest model demonstrated strong predictive performance in both scenarios.
Conclusions:
- Machine learning can feasibly predict long-term CPAP adherence in OSA patients.
- Incorporating early adherence metrics substantially enhances predictive accuracy.
- Preliminary results suggest a potential for personalized follow-up and resource optimization, requiring further validation.
