Automated interpretation of scatter signatures aimed at tissue morphology identification

Abstract

An automated algorithm and methodology is presented to pathologically classify the scattering changes encountered in the raster scanning of normal and tumor pancreatic tissues using microsampling reflectance spectroscopy. A quasiconfocal reflectance imaging system was used to directly measure the tissue scatter reflectance in situ, and the spectrum was used to identify the scattering power, amplitude and total wavelength-integrated intensity. Pancreatic tumor and normal samples were characterized using the instrument and subtle changes in the scatter signal were encountered within regions of each sample. Discrimination between normal vs. tumor tissue was readily performed using an Artificial Neural Network (ANN) classifier algorithm. A similar approach has worked also for regions of tumor morphology when statistical pre-processing of the scattering parameters was included to create additional data features. This automated interpretation methodology can provide a tool for guiding surgical resection in areas where microscopy imaging do not reach enough contrast to assist the surgeon.