Cytological image analysis with a genetic fuzzy finite state machine

摘要: The objective of this research is to design a pattern recognition system based on Fuzzy Finite State Machine (FFSM). We try find an optimal FFSM with Genetic Algorithms (GA). In order validate system, the classifier has been applied real problem: distinction between normal and abnormal cells in cytological breast fine needle aspirate images peritoneal fluid images. characteristic used discrimination texture measurement chromatin distribution cellular nuclei. Furthermore, effectiveness method as compared other existing supervised unsupervised methods evaluated Receiver Operating Curves (ROC) methodology.