Using Evolved Artificial Neural Networks for Providing an Emergent Segmentation with an Active Net Model
A novel segmentation method using deformable models for medical image segmentation was developed. As deformable model, Topological Active Nets (TAN) were used, model which integrates features of region-based and boundary-based segmentation techniques. The model deformation through time is controlled by an Artificial Neural Network (ANN) that learns how to move the nodes of the model based on their energy surrounding. The ANN is applied to each of the nodes and in different temporal steps until the final segmentation is reached. The ANN training is obtained by simulated evolution, using Differential Evolution (DE) to automatically obtain the ANN that provides the emergent segmentation. The proposed methodology was adapted and tested in three different medical domains, that is, Computed Tomography (CT), Cone Beam Computed Tomography (CBCT) and eye fundus images to demonstrate the potential of the segmentation technique.
KeywordsTopological Active Nets Differential Evolution Artificial Neural Networks Medical Imaging
This paper has been partly funded by the Ministry of Science and Innovation through grant contracts TIN2011-25476 and TIN2011-27294 and by the Consellería de Industria, Xunta de Galicia, through grant contract 10/CSA918054PR.
- 1.Tsumiyama, K., Yamamoto, K.: Active net: active net model for region extraction. IPSJ SIG Notes 89(96), 1–8 (1989)Google Scholar
- 3.Ansia, F., Penedo, M., Mariño, C., Mosquera, A.: A new approach to active nets. Pattern Recognit. Image Anal. 2, 76–77 (1999)Google Scholar
- 4.Williams, D., Shah, M.: A fast algorithm for active contours and curvature estimation. CVGIP Image Underst. 55(1), 14–26 (1992)Google Scholar
- 9.Zitzler, E., Laumanns, M., Thiele, L.: SPEA2: Improving the strength pareto evolutionary algorithm. In: EUROGEN: Evolutionary Methods for Design. Optimisation, and Control. 2002, 95–100 (2001)Google Scholar
- 11.Novo, J., Penedo, M.G., Santos, J.: Multiobjective optimization of the 3D topological active volume segmentation model. In: ICAART: International Conference on Agents and. Artificial Intelligence. 2011, 236–241 (2011)Google Scholar
- 14.Price, K., Storn, R., Lampinen, J.: Differential evolution. A Practical Approach to Global Optimization (Natural Computing Series). Springer, New York (2005)Google Scholar
- 15.Feoktistov, V.: Differential Evolution: In Search of Solutions. Springer, New York (2006)Google Scholar
- 18.VARIA: VARPA Retinal Images for Authentication. Varpa Web site at. http://www.varpa.es/varia.html (2008)