Prediction of Periventricular Leukomalacia Occurrence in Neonates Using a Novel Unsupervised Learning Method

Prior work has documented that Support Vector Machine (SVM) classifiers can be powerful tools in predicting clinical outcomes of complex diseases such as Periventricular Leukomalacia (PVL). A preceding study indicated that SVM performance can be improved significantly by optimizing the supervised training set used during the learning stage of the overall SVM algorithm. This preliminary work, as well as the complex nature of the PVL data suggested integration of the active learning algorithm into the overall SVM framework. The present study supports this initial hypothesis and shows that active learning SVM type classifier performs considerably well and outperforms normal SVM type classifiers when dealing with clinical data of high dimensionality.