Automatic segmentation of molecular pathology images using a robust mixture model with Markov random fields
The segmentation of molecular pathology images is important for the assessment of clinical behaviour of disease conditions. We consider a robust mixture model-based approach to segment pathology images into different tissue components, with the use of Markov random fields to account for the spatial continuity of image intensities. Segmentation and estimation of tissue parameters quantify the size of various tissue components and can be used to assess progression of disease or to evaluate effect of drug therapy. The method is illustrated using simulated data and pathology images of cancer patients.
2013 International Conference on Digital Image Computing: Techniques and Applications (DICTA)
Medical and Health Sciences not elsewhere classified