Development of a Neuro-fuzzy MR Image Segmentation Approach Using Fuzzy C-Means and Recurrent Neural Network
A neuro-fuzzy clustering framework has been presented for a meaningful segmentation of Magnetic Resonance medical images. MR imaging provides detail soft tissue descriptions of the target body object and it has immense importance in today’s non-invasive therapeutic planning and diagnosis methods. The unlabeled image data has been classified using fuzzy c-means approach and then the data has been used for training of an Elman neural network. The trained neural net is then used as a ready-made tool for MRI segmentation.
KeywordsMedical Image Segmentation Neuro-Fuzzy Elman Recurrent Neural Network (ERNN)
- 1.Ray, D., Dutta Majumder, D.: Studies on Some multimodal medical image registration approaches for diagnostics and therapeutic planning: With some case studies. Science Letters, NAS 28(5&6), 135–154 (2005)Google Scholar
- 3.Bezdek, J.C., Hathaway, R.J., Sabin, M.J., Tucker, W.T.: Convergence theory for fuzzy C-means: counter examples and repairs. IEEE Trans. Syst. Man Cybern. 17873–17877 (1987)Google Scholar
- 6.Powell, M.J.D.: Radial basis functions for multivariable interpolation: a review. In: Algorithms for Approximation, pp. 143–167. Clarendon Press, Oxford (1987)Google Scholar