J Jelonek | PubMed Health

PubMed1999

Intelligent decision support in pathomorphology.

This paper presents a novel approach to computer-supported diagnosing based on microscopic images of histological sections. A method of extraction of textural feature is presented, which is in a sense complementary to the texture-based segmentation. The textural feature is obtained by tracing the process of image segmentation. For classification, a n2-classifier oriented to multi-class problems has been used. The paper presents also an empirical verification of the proposed approach on 700 microscopic images representing 14 classes of CNS neuroepithelial tumours, in which case an encouraging accuracy of classification on the testing set (70.6%) has been obtained.

Central Nervous System Neoplasms↗

PubMed1997

Feature subset selection for classification of histological images.

Classification of histological images is considered in this paper. The task is to distinguish different classes of tumours of the central nervous system on the basis of features extracted from microscopic slides. The number of extracted features is relatively high and some of them seem to be irrelevant for classification of the images. Thus, the main objective of this study is to select such a feature subset that improves the predictive accuracy of the classifier. The wrapper approach is chosen to obtain this aim, where a search for the good subset of features is made using a non-parametric case-base classifier. To guide a search process, a forward beam selection algorithm is introduced. It sequentially adds relevant features in a parallel way for the most promising subsets. It is shown that the proposed approach gives good predictive accuracy for the considered histopathological problem.

Algorithms↗