An Evolutionary Misclassification Cost Minimization Approach for Medical Diagnosis

Abstract

This chapter illustrates how a misclassification cost matrix can be incorporated into an evolutionary classification system for medical diagnosis. Most classification systems for medical diagnosis have attempted to minimize the misclassifications (or maximize correctly classified cases). The minimizing misclassification approach assumes that Type I and Type II error costs for misclassification are equal. There is evidence that these costs are not equal and incorporating costs into classification systems can lead to superior outcomes. We use principles of evolution to develop and test a genetic algorithm (GA) based approach that incorporates the asymmetric Type I and Type II error costs. Using simulated and real-life medical data, we show that the proposed approach, incorporating Type I and Type II misclassification costs, results in lower misclassification costs than LDA and GA approaches that do not incorporate these costs.