Comparison and hybridization of neural networks and fuzzy logic in biomedical applications

Neural networks and fuzzy logic are cousins under the aegis of human-thought-inspired soft computing, which, in contrast to tradition or hard computing, is robust to imprecision, uncertainty, partial truth, and approximation. As such, these two techniques share common abilities; however, they are very different. Individually, both techniques are widely used in biomedical applications, but there is a real power in combining the two to form neurofuzzy or fuzzy-neural systems (not the same thing). In addition to expounding neural networks and fuzzy logic individually and in hybrids, this paper presents biomedical applications of these soft computing approaches.