Sleep apnea detection using Xception and residual network

Nida Nasir; Feras Barneih; Omar Alshaltone; Mohammad AlShabi; Talal Bonny; Ahmed Al Shammaa

doi:10.1117/12.2664009

14 June 2023 Sleep apnea detection using Xception and residual network

Nida Nasir, Feras Barneih, Omar Alshaltone, Mohammad AlShabi, Talal Bonny, Ahmed Al Shammaa

Proceedings Volume 12548, Smart Biomedical and Physiological Sensor Technology XX; 125480K (2023) https://doi.org/10.1117/12.2664009
Event: SPIE Defense + Commercial Sensing, 2023, Orlando, Florida, United States

Conference Poster

Abstract

Sleep apnea is a disorder that has the potential to be life-threatening, that is characterized by irregular breathing patterns. In order to improve the diagnosis and prediction of sleep apnea, a study was conducted to develop a high-accuracy detection method using machine learning. This method involved the use of a convolutional neural network classifier, which was trained using public data sets of ECG signals from both apnea patients and healthy volunteers. The CNN model was able to attain a level of accuracy of 94.12% using the Xception model and 91.18% using the ResNet50 model. According to the study’s findings, using deep learning techniques can be a helpful strategy to enhance sleep apnea diagnosis and prediction.

Citation Download Citation

Nida Nasir, Feras Barneih, Omar Alshaltone, Mohammad AlShabi, Talal Bonny, and Ahmed Al Shammaa "Sleep apnea detection using Xception and residual network", Proc. SPIE 12548, Smart Biomedical and Physiological Sensor Technology XX, 125480K (14 June 2023); https://doi.org/10.1117/12.2664009

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available