Conte, L.; De Nunzio, G.; Giombi, F.; Lupo, R.; Arigliani, C.; Leone, F.; Salamanca, F.; Petrelli, C.; Angelelli, P.; De Benedetto, L.; Arigliani, M. Machine Learning Models to Enhance the Berlin Questionnaire Detection of Obstructive Sleep Apnea in at-Risk Patients. Appl. Sci.2024, 14, 5959.
Conte, L.; De Nunzio, G.; Giombi, F.; Lupo, R.; Arigliani, C.; Leone, F.; Salamanca, F.; Petrelli, C.; Angelelli, P.; De Benedetto, L.; Arigliani, M. Machine Learning Models to Enhance the Berlin Questionnaire Detection of Obstructive Sleep Apnea in at-Risk Patients. Appl. Sci. 2024, 14, 5959.
Conte, L.; De Nunzio, G.; Giombi, F.; Lupo, R.; Arigliani, C.; Leone, F.; Salamanca, F.; Petrelli, C.; Angelelli, P.; De Benedetto, L.; Arigliani, M. Machine Learning Models to Enhance the Berlin Questionnaire Detection of Obstructive Sleep Apnea in at-Risk Patients. Appl. Sci.2024, 14, 5959.
Conte, L.; De Nunzio, G.; Giombi, F.; Lupo, R.; Arigliani, C.; Leone, F.; Salamanca, F.; Petrelli, C.; Angelelli, P.; De Benedetto, L.; Arigliani, M. Machine Learning Models to Enhance the Berlin Questionnaire Detection of Obstructive Sleep Apnea in at-Risk Patients. Appl. Sci. 2024, 14, 5959.
Abstract
Objective: With just ten questions, the Berlin questionnaire (BQ) stands out as one of the simplest and most widely implemented non-invasive screening tools for detecting subjects at high risk for Obstructive Sleep Apnea (OSA), a still underdiagnosed syndrome characterized by partial or complete obstruction of the upper airways during sleep. The main aim of this study was to enhance the diagnostic accuracy of the BQ through Machine Learning (ML) techniques. Methods: A ML classifier (hereafter, ML-10) was trained using the ten questions of the standard BQ. A simplified variant of the BQ, BQ-2, which comprises only two questions out of the total of ten, was also assessed in a ML context. A 10-fold cross validation scheme was used. Ground truth was provided by the Apnea-Hypopnea Index (AHI) measured by Home Sleep Apnea Testing. Model performance was determined comparing ML-10 and BQ-2 with the standard BQ by the Receiver Operating Characteristic Curve (ROC), Area Under the Curve (AUC), sensitivity, specificity, and accuracy. Results: ML-10 demonstrated superior performance in predicting the risk for OSA compared to the standard BQ and was also capable in classifying OSA with two different AHI thresholds (AHI>=15, AHI>=30), typically used in clinical practice. Remarkably, BQ-2 was also better in sensibility to assess moderate to severe OSA (AHI≥ 15) compared to the ML-10. Conclusions: The study underscores the importance of integrating ML techniques for early OSA detection, suggesting a direction for future research to improve diagnostic processes and patient outcomes in sleep medicine.
Keywords
Obstructive Sleep Apnea; OSA; berlin questionnaire; Machine Learning; artificial intelligence
Subject
Engineering, Bioengineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.