Bibliographic details

Journal: Neurocomputing
Status: in press

About the authors

Francesco Bianconi* is a Lecturer in the Department of Engineering at the University of Perugia, Italy.
E-mail: bianco@ieee.org

Alberto Álvarez-Larrán is a consultant Haematologist in the Haemathology Department at the Hospital del Mar, Barcelona, Spain,
and Associate Professor at the Autonomous University of Barcelona.
E-mail: 95967@parcdesalutmar.cat

Antonio Fernández is a Senior Lecturer in the School of Industrial Engineering at the University of Vigo, Spain.
E-mail: antfdez@unipg.it



*Corresponding author

Abstract

In this work we propose the use of image features based on visual perception for discriminating epithelium and stroma in histological images. In particular, we assess the capability of the following five visual features to correctly discriminate epithelium from stroma in digitised tissue micro-arrays of colorectal cancer: coarseness, contrast, directionality, line-likeliness and roughness. The use of features directly related to human perception makes it possible to evaluate the tissue's appearance on the basis of a set of meaningful parameters; moreover, the number of features used to discriminate epithelium from stroma is very small. In the experiments we used histologically-verified, well-defined images of epithelium and stroma to train three classifiers based on Support Vector Machines (SVM), Nearest Neighbour rule (1-NN) and Naive Bayes rule (NB.) We optimised SVM's parameters on a validation set, and estimated the accuracy of the three classifiers on a independent test set. The experiments demonstrate that the proposed features can correctly discriminate epithelium from stroma with state-of-the-art accuracy.





DISCLAIMER
The information and content on this Web site are provided with no warranty whatsoever. Any use for scientific or any other purpose is conducted at your own risk and under your own responsibility. The authors are not liable for any damages, including any consequential damages, of any kind that may result to the user from the use of the materials on this Web site or of any of the products or services described hereon.