Design of auto-classifying system and its application in Raman spectroscopy diagnosis of gastric carcinoma

A tentative user-friendly auto-classifying system is developed to classify Raman spectra of gastric carcinoma tissues. By combining two efficient fuzzy clustering methods: artificial neural network (ANN) and fuzzy C-means (FCM), the software system integrates high robustness of ANN and strong quantity estimation ability of FCM, and provides more credible classifying results for the Raman spectroscopy auto-diagnosis of gastric carcinoma. Also the system can be applied to classifying of other spectroscopy by some necessary alteration of feature vector.