Apnea Event Detection Using Machine Learning Technique for the Clinical Diagnosis of Sleep Apnea Syndrome

A sort of rest problem that is portrayed by breaks in breathing or examples of shallow or phenomenal breathing during rest is known as sleep apnea. Side effects of sleep apnea may happen when a patient is not having any admittance to a doctor. Absence of access is the main source of information misfortune which is fundamental and basic in conclusion and treatment. Since the relationship between ECG sign and sleep apnea issue has been demonstrated and as per the way that other bio-signal cannot be recorded in the method of recording Holter ECG. In this examination, sleep apnea ages were distinguished by means of managed ECG information handling. The information is gathered from Physio net ATM. The detection method can be acquired by using time, frequency, and statistical features of Electrocardiogram signals. Apnea event detection using the extracted features is evaluated through different classifiers. Ensemble-Bagged Tree classifier has shown a higher accuracy of 89.6% with MIT-BIH Apnea datasets than that of decision tree, support vector machine, and K-nearest neighborhood classifiers.