An untrained and unsupervised method for MRI brain tumor segmentation
Autor: | Tom Haeck, Paul Suetens, Frederik Maes |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2016 |
Předmět: |
Level set (data structures)
business.industry Computer science Brain tumor Scale-space segmentation Pattern recognition Image segmentation medicine.disease computer.software_genre 01 natural sciences 030218 nuclear medicine & medical imaging Intensity (physics) 010101 applied mathematics 03 medical and health sciences 0302 clinical medicine PSI_MIC Voxel medicine Computer vision Segmentation Artificial intelligence 0101 mathematics business computer |
Zdroj: | ISBI |
Popis: | © 2016 IEEE. We present a fully-automated MRI brain tumor segmentation method that does not require any manually annotated training data. The method is independent of the scanner or acquisition protocol and is directly applicable to any individual patient image. An Expectation Maximization-approach is used to estimate intensity models for both normal and tumorous tissue. The segmentation is represented by a level-set that is iteratively updated to label voxels as normal or tumorous, based on which intensity model explains the voxels' intensity the best. The method is compared with the method by Menze et al. [1], which is considered to be a benchmark for unsupervised tumor segmentation. The performance of our method for segmenting the tumor volume is summarized by an average Dice score of 0.87 ± 0.06 on the training data set of the MICCAI BraTS Challenge 2012-2013. Haeck T., Maes F., Suetens P., ''An untrained and unsupervised method for MRI brain tumor segmentation'', Proceedings 13th IEEE international symposium on biomedical imaging - ISBI 2016, pp. 265-268, April 13-16, 2016, Prague, Czech Republic. ispartof: pages:265-268 ispartof: Proceedings ISBI 2016 vol:2016-June pages:265-268 ispartof: IEEE international symposium on biomedical imaging - ISBI 2016 location:Prague, Czech Republic date:13 Apr - 16 Apr 2016 status: published |
Databáze: | OpenAIRE |
Externí odkaz: |