A Gravitational Search Algorithm using Fuzzy Adaptation of Parameters for Optimization of Ensemble Neural Networks in Medical Imaging

摘要

In this paper we describe a gravitational search algorithm (GSA) using type-2 fuzzy logic for the optimization of ensemble neural networks. The GSA is based on populations, and uses physical rules: the law from gravity and law of motion. This method is applied to a pattern recognition problem in medical imaging. We are using echocardiographic medical images, for disease and healthy patients. The medical images can be used in methods for detection of diseases by physicians.