Backpropagation Neural Network for Analysis and Classification of Fluorescence Spectroscopy of Squamous Cell Carcinoma in Animal Model

The present study aims to evaluate the performance of a backpropagation neural network (BPNN) using the principal component analysis (PCA) of fluorescence spectra for discrimination between normal skin and skin tumor on mice. The fluorescence spectra were acquired from nude mice with induced squamous cell carcinoma (SCC). The artificial neural network (ANN) used in this study is a classical multiplayer feed-forward type with a back-propagation algorithm. The classification results show this technique as promising for healthy and unhealthy tissue classification. During the validation, the network classified 100% of the training set spectra and 90% of the test set.