Indrajit Mandal and Sairam.N, Sastra University, India

Improving accuracy of supervised classification algorithms in biomedical applications, especially CADx, is one of active area of research. This paper proposes construction of rotation forest (RF) ensemble using 20 learners over two clinical datasets namely lymphography and backache. We propose a new feature selection strategy based on support vector machines optimized by particle swarm optimization for relevant and minimum feature subset for obtaining higher accuracy of ensembles. We have quantitatively analyzed 20 base learners over two datasets and carried out the experiments with 10 fold cross validation leave-one-out strategy and the performance of 20 classifiers are evaluated using performance metrics namely accuracy (acc), kappa value (K), root mean square error (RMSE) and area under receiver operating characteristics curve (ROC). Base classifiers succeeded 79.96% & 81.71% average accuracies for lymphography & backache datasets respectively. As for RF ensembles, they produced average accuracies of 83.72% & 85.77% for respective diseases. The paper presents promising results using RF ensembles and provides a new direction towards construction of reliable and robust medical diagnosis systems.

Rotation forest, support vector machines, particle swarm optimization, ensembles, medical diagnosis systems.