Optimization of Sleep Apnea Detection using SpO2 and ANN

Repetitive respiratory disturbance during sleep is called Sleep Apnea Hypopnea Syndrome and causes various diseases. Different features and classifiers have been used by different researchers to detect sleep apnea. This study is undertaken to identify the better performing blood oxygen saturation features subset using an Artificial Neural Network classifier for sleep Apnea detection. A database of 8 subjects with one-minute annotation is used to test the proposed system. The optimized system has seven features chosen from a total set of sixty-one features presenting a high accuracy rate using a genetic algorithm. Artificial Neural Network was able to achieve 97.7 percentage of accuracy with only seven features chosen by the Genetic algorithm.