Texture analysis

Promising results from the technical development image analysis and segmenting algorithms for Nucleotyping, have now made it possible to assess automatic methods for classifying cells and tissue in routine pathology sections.

Overview

Combined with traditional graph theory and some of our own ideas, we believe it should be possible to develop a system for making some types of histopathological classification automatic.

We have demonstrated that we are able to identify nuclei in HE stained routine specimens from prostate cancer, and achieved promising classification results related to Gleason grading.