Obstructive sleep apnea detection from ECG signal using neuro-fuzzy classifier

Obstructive sleep apnea syndrome (OSAS) is a common problem defined by frequent process of breathing due to the partial or complete obstruction of upper airway for short period during sleep. This leads to serious cardiovascular and other problems or even death if not properly detected. The existing method of detection considers temporal dependency within ECG segmented signal. Here HMM and corresponding state transmission probability, observed probability and initial state probability were used to detect the condition. However the system lacks proper feature selection criteria and also fails in detecting the sleep apnea. To overcome, a T-test method has been proposed to select the feature from RR interval and QRS complex. Also to reduce the miss classification a Neuro Fuzzy detection method has been introduced. A neuro fuzzy system is a fuzzy learning that uses learning algorithm derived from neural network theory. The required ECG signal is taken from Physionet database. The artificial intelligence based detection method reduce false positive rate in analysis of sleep apnea.