Send to

Choose Destination
J Med Syst. 2016 Dec;40(12):282. Epub 2016 Oct 27.

New Rule-Based Algorithm for Real-Time Detecting Sleep Apnea and Hypopnea Events Using a Nasal Pressure Signal.

Author information

Interdisciplinary Consortium on Advanced Motion Performance, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, TX, USA.
Department of Biomedical Engineering, Yonsei University, Wonju, Republic of Korea.
Division of Pulmonary and Critical Care Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea.
Department of Biomedical Engineering, Yonsei University, Wonju, Republic of Korea.


We developed a rule-based algorithm for automatic real-time detection of sleep apnea and hypopnea events using a nasal pressure signal. Our basic premise was that the performance of our new algorithm using the nasal pressure signal would be comparable to that using other sensors as well as manual annotation labeled by a technician on polysomnography study. We investigated fifty patients with sleep apnea-hypopnea syndrome (age: 56.8 ± 10.5 years, apnea-hypopnea index (AHI): 36.2 ± 18.1/h) during full night PSG recordings at the sleep center. The algorithm was comprised of pre-processing with a median filter, amplitude computation and apnea-hypopnea detection parts. We evaluated the performance of the algorithm a confusion matric for each event and statistical analyses for AHI. Our evaluation achieved a good performance, with a sensitivity of 86.4 %, and a positive predictive value of 84.5 % for detection of apnea and hypopnea regardless of AHI severity. Our results indicated a high correlation with the manually labeled apnea-hypopnea events during PSG, with a correlation coefficient of r = 0.94 (p < 0.0001) and a mean difference of -2.9 ± 11.6 per hour. The proposed new algorithm could provide significant clinical and computational insights to design a PSG analysis system and a continuous positive airway pressure (CPAP) device for screening sleep quality related in patients with sleep apnea-hypopnea syndrome.


Apnea; Hypopnea; Nasal pressure signal; Real-time detection

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center