DocumentCode :
3146355
Title :
Procedure to identify sleep apnea events from statistical features
Author :
Tan-a-ram, S. ; Thanawattano, Chusak
Author_Institution :
Biomed. Signal Process. Lab., Nat. Electron. & Comput. Technol. Center, Pathumthani, Thailand
Volume :
3
fYear :
2010
fDate :
16-18 Oct. 2010
Firstpage :
996
Lastpage :
1001
Abstract :
This paper proposes the methods for identifying time period of sleep apnea events with several statistic values. Firstly, we calculate the RR-interval values from electrocardiography (ECG) signal. Then we compute several statistic values, e.g. mean, standard deviation and coefficient of variation from the calculated RR-interval values. In this research, we consider both the large number of sleep apnea case and the small number of sleep apnea case and then propose the several conditions for identifying time period of sleep apnea events correspond to both case. These conditions are derived from our experiment with data set from standard apnea database. Finally, the detection results shown that the accuracy of our proposed method with the large number of sleep apnea case is between 60% - 88% and average of the accuracy is 71.03% for the proposed method based on condition 3 that can be comparable with one of reference paper, but in the small number of sleep apnea case our method has the accuracy of our proposed method close to 100% which show that our method is very efficient in detecting time period of sleep apnea compared with result from reference paper and annotation file from database.
Keywords :
electrocardiography; medical signal detection; medical signal processing; sleep; statistical analysis; ECG; RR interval values; electrocardiography; sleep apnea event detection; statistical features; Accuracy; Computer aided software engineering; Databases; Electrocardiography; Sleep apnea; Training data; coefficient of variation; mean; sleep apnea; standard deviation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Engineering and Informatics (BMEI), 2010 3rd International Conference on
Conference_Location :
Yantai
Print_ISBN :
978-1-4244-6495-1
Type :
conf
DOI :
10.1109/BMEI.2010.5639733
Filename :
5639733
Link To Document :
بازگشت