Detection of obstructive sleep apnea from cardiac internet interval time series

Reference:
Mietus JE, Peng CK, Ivanov PC, Goldberger AL. Detection of obstructive sleep apnea from cardiac internet interval time series. DOI:10.1109/CIC.2000.898634

Objectives:
Present a new automated method to diagnose and quantify obstructive sleep apnea from single-lead electrocardiograms based on the detection of the periodic oscillations in cardiac intervals that are often associated with prolonged cycles of sleep apnea.

Conclusions:
The algorithm correctly classified 28 out of 30 cases (93.3%) of both sleep apnea and normal subjects, and correctly identified the presence or absence of sleep apnea in 14,591 out of a total of 17,268 minutes (84.5%) of the data from the test set.

Practical Significance:
This was the first work by some of the co-inventors of what became Cardiopulmonary Coupling, that was first published five years after this competition took place, called Computers in Cardiology.

View Publication

Detection of obstructive sleep apnea from cardiac internet interval time series