Vol. 230, Issue 2, February 2019, pp. 7-13

Abstract: Obstructive sleep apnea (OSA) is defined as a sleep disorder caused by periodic obstruction or stricture of the upper airway. It was recently discovered that OSA is associated with cardiovascular diseases. Since polysomnography, which was previously used to evaluate OSA, requires high-cost and has time/spatial restrictions, HRV analysis and acceleration sensors are being investigated as new parameters of OSA. In this study, ECG electrodes were attached to the acceleration sensors to simultaneously measure 2 different types of biomarkers: acceleration of thoracic motion and ECG. HRV was measured using the ECG and it was analyzed to obtain LF/HF ratios, indicators of autonomic nervous systems, at different times. Also, ECG Fourier coefficient was used. A filter was used to remove the noise from the acceleration data and 6 characteristic points of OSA were detected. The 6 characteristic points of OSA were analyzed using the Weka program. Adaboost, one of the boosting algorithms, was used. 10-folds cross validation method was used to evaluate the model, which showed a sensitivity of 90 %, specificity of 80 %, accuracy of 85 % in diagnosing OSA.

Keywords: Computer-aided Diagnosis, Sleep Apnea, Adaboost, Machine learning.

Click <here> or title of paper to download the full pages article in pdf format

This work is licensed under a Creative Commons 4.0 International License

1999 - 2019 Copyright ©, International Frequency Sensor Association (IFSA) Publishing, S.L. All Rights Reserved.

Home - News - Links - Archives - Tools - Voltage-to-Frequency Converters - Standardization - Patents - Marketplace - Projects - Wish List - e-Shop - Sensor Jobs - Membership - Videos - Publishing - Site Map - Subscribe - Search

 Members Area -Sensors Portal -Training Courses - S&T Digest - For advertisers - Bookstore - Forums - Polls - Submit Press Release - Submit White Paper - Testimonies - Twitter - Facebook - LinkedIn