A Novel Approach for Cardiotocography Paper Digitization and Classification for Abnormality Detection

Cardiotocography (CTG) is a clinical procedure that used to track and gauge the severity of fetal distress. Although CTG most often equipment monitor assess health fetus, high rate false positive results due visual interpretation significantly contributes needless surgical delivery or delayed intervention. In this study, novel approach introduced where both printing paper digitized machine learning employed detect abnormality in signal. Image processing-based preprocessing steps are make more convenient extract Various signal-processing techniques calibrate extracted Then, Empirical Mode Decomposition (EMD) decompose signal into its frequency components instantaneous spectral entropy features extracted. After feature normalization selection with ReliefF algorithm, support vector machines (SVM) for classification normal abnormal classes. A dataset experimental works various performance evaluation metrics achievement proposed method. 10-fold cross-validation-based experiments show method quite efficient detection papers an average accuracy score around 90.0% produced.