Mumford-Shah based unsupervised segmentation of brain tissue on MR images

Date

2014-01-01

Author

Cevik, A.
Eyüboğlu, Behçet Murat

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

21
views

0
downloads

Cite This

Automated segmentation of different tissues on medical images is a crucial concept for medical image analysis. In this study, unsupervised image segmentation problem is generalized as a Mumford-Shah energy minimization problem, and several solution proposals for the problem are investigated. Ambrosio-Tortorelli approximation method is implemented, and the performance of the algorithm on magnetic resonance (MR) images of brain is evaluated. First image used in the experiments is chosen among the ones which contain an edema formation due to a brain tumor, and the second one belongs to a healthy subject on which gray matter/white matter segmentation is aimed. Acquired results are presented in visual, tabular and numerical forms. Results and performance are discussed and quantitatively evaluated. © Springer International Publishing Switzerland 2014.

Subject Keywords

Brain tumor segmentation, Gray matter/white matter segmentation, Medical image processing, Unsupervised image segmentation

URI

https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84891319079&origin=inward
https://hdl.handle.net/11511/107993

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar