Pairwise Classifier Approach to Automated Diagnosis of Disorder Degree of Obstructive Sleep Apnea Syndrome: Combining of AIRS and One versus One (OVO-AIRS)

摘要

Artificial Immune Recognition System (AIRS) is an immune inspired supervised classification algorithm and also works in classifying of multi class datasets. But the performance of AIRS classifier in classifying multi class datasets is generally lower than its performance in case of classifying two class datasets. In order to overcome this problem, we have combined the one-versus-one (OVO) and AIRS in the diagnosis of disorder degree of obstructive sleep apnea syndrome (OSAS) that affects both the right and the left cardiac ventricle. The OSAS dataset consists of four classes including of normal (25 subjects), mild OSAS (AHI (Apnea Apnea and Hypoapnea Index)=5-15 and 14 subjects), moderate OSAS (AHI<15-30 and 18 subjects), and serious OSAS (AHI>30 and 26 subjects). In the extracting of features that is characterized the OSAS disease, the clinical features obtained from Polysomnography used diagnostic tool for obstructive sleep apnea in patients clinically suspected of suffering from this disease have been used. The used clinical features are Arousals Index (ARI), Apnea and Hypoapnea Index (AHI), SaO2 minimum value in stage of REM, and Percent Sleep Time (PST) in stage of SaO2 intervals bigger than 89%. We have used two fold cross validation to split OSAS dataset and also used the classification accuracy, sensitivity- specificity analysis, and confusion matrix to evaluate the performance of proposed method. While AIRS algorithm obtained 90.24% classification accuracy, the proposed method based on AIRS algorithm and OVO achieved 98.24% classification accuracy. These results show that the proposed method can confidently be used in the determining of disorder degree of OSAS.