Classification of Sleep Apnea Events Using Nasal Air Flow (NAF) Signals.

Authors

Fatma Abdel-Meguid, Electronics and Communications Engineering Department., Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.
Fatma Abou-Chadi, Electronics and Communications Engineering Department., Faculty o Engineering., El-Mansoura University., Mansoura., Egypt.
S. Loza, Cairo Center for Sleep Disorder., Cairo., Egypt.

Corresponding Author

Abdel-Meguid, Fatma

Subject Area

Electronics and Communication Engineering

Article Type

Original Study

Abstract

Detecting and diagnosing different types of Sleep Apnea (Obstructive, Central and Hypopneaisone of the major tasks in sleep medicine. Clinically, analyzing Nasal Airflow (NAF) signal is the most sufficient and direct reliably effective method for the automatic detection of Sleep Apnea events by checking the airflow amplitude reduction. This paper investigates and compares the performance of three classifiers: a kohen self-organizing map (SOM), Adaptive Neuro Fuzzy Inference System ?(ANFIS) AND Hidden Markov Model (HMM) for the classification of Sleep apnea events. The results have shown that the highest correct classification rate is 95.7% when using HMM with order 13.

Recommended Citation

Abdel-Meguid, Fatma; Abou-Chadi, Fatma; and Loza, S. (2021) "Classification of Sleep Apnea Events Using Nasal Air Flow (NAF) Signals.," Mansoura Engineering Journal: Vol. 37 : Iss. 4 , Article 12.
Available at: https://doi.org/10.21608/bfemu.2021.199195