Automatic evolutionary medical image segmentation using deformable models

Abstract : This paper describes a hybrid level set approach to medical image segmentation. The method combines region-and edge-based information with the prior shape knowledge introduced using deformable registration. A parameter tuning mechanism, based on Genetic Algorithms, provides the ability to automatically adapt the level set to different segmentation tasks. Provided with a set of examples, the GA learns the correct weights for each image feature used in the segmentation. The algorithm has been tested over four different medical datasets across three image modalities. Our approach has shown significantly more accurate results in comparison with six state-of-the-art segmentation methods. The contributions of both the image registration and the parameter learning steps to the overall performance of the method have also been analyzed.
Complete list of metadatas

Cited literature [30 references]  Display  Hide  Download

https://hal.inria.fr/hal-01221343
Contributor : Pablo Mesejo Santiago <>
Submitted on : Tuesday, October 27, 2015 - 6:09:29 PM
Last modification on : Tuesday, April 30, 2019 - 3:42:27 PM
Long-term archiving on : Friday, April 28, 2017 - 5:36:17 AM

File

paperFinal.pdf
Files produced by the author(s)

Identifiers

Citation

Andrea Valsecchi, Pablo Mesejo, Linda Marrakchi-Kacem, Stefano Cagnoni, Sergio Damas. Automatic evolutionary medical image segmentation using deformable models. 16th IEEE Congress on Evolutionary Computation (CEC’14), Jul 2014, Beijing, China. pp.97-104, ⟨10.1109/CEC.2014.6900466⟩. ⟨hal-01221343⟩

Share

Metrics

Record views

473

Files downloads

327