Application of rough set based dynamic parameter optimization to MRI segmentation
Sebastian Widz, Dominik Ślȩzak, K. Revett · IEEE Annual Meeting of the Fuzzy Information, 2004. Processing NAFIPS '04. · 2004
We introduce a multi-spectral MRI segmentation technique based on approximate reducts derived from the data mining paradigm of the theory of rough sets. We use genetic algorithms to tune the parameterized attributes and search for the best segmentation models based on approximate reducts.