Class prediction of cancer using probabilistic neural networks and relative correlation metric

Abstract

Accurate diagnosis and classification is the key issue for the optimal treatment of cancer patients. Several studies demonstrate that cancer classification can be estimated with high accuracy, sensitivity, and specificity from microarray-based gene expression profiling using artificial neural networks. In this paper, a comprehensive study was undertaken to investigate the capability of the probabilistic neural networks along with a feature selection method in the application of cancer classification. The feature selection method is based on the correlation with the class distinction. The experimental results show that the conjugation of the probabilistic neural network and the feature selection method can achieve 100% recognition accuracy in the ALL/AML classification, and also attain satisfactory results in two colon cancer data sets.