Patch-based generative shape model and MDL model selection for statistical analysis of archipelagos

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

We propose a statistical generative shape model for archipelago-like structures. These kind of structures occur, for instance, in medical images, where our intention is to model the appearance and shapes of calcifications in x-ray radio graphs. The generative model is constructed by (1) learning a patch-based dictionary for possible shapes, (2) building up a time-homogeneous Markov model to model the neighbourhood correlations between the patches, and (3) automatic selection of the model complexity by the minimum description length principle. The generative shape model is proposed as a probability distribution of a binary image where the model is intended to facilitate sequential simulation. Our results show that a relatively simple model is able to generate structures visually similar to calcifications. Furthermore, we used the shape model as a shape prior in the statistical segmentation of calcifications, where the area overlap with the ground truth shapes improved significantly compared to the case where the prior was not used.
Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging : First International Workshop, MLMI 2010, Held in Conjunction with MICCAI 2010, Beijing, China, September 20, 2010. Proceedings
EditorsFei Wang, Pingkun Yan, Kenji Suzuki, Dinggang Shen
Number of pages8
Publication date2010
ISBN (Print)978-3-642-15947-3
ISBN (Electronic)978-3-642-15948-0
Publication statusPublished - 2010
Event1st International Workshop on Machine Learning in Medical Imaging - Beijing, China
Duration: 20 Sep 201020 Sep 2010
Conference number: 1


Conference1st International Workshop on Machine Learning in Medical Imaging
SeriesLecture notes in computer science

ID: 170194357