Multichannel segmentation of magnetic resonance cerebral images based on neural networks

In this article, we present an approach for the segmentation of magnetic resonance images of the brain, based on a Hopfield neural network. We formulate the segmentation problem as a minimization of an energy function constructed with two terms, the cost-term, that is a sum of errors´ squares, and the second term is a temporary noise added to the cost-term as an excitation to the network to escape from certain local minimums and be more close to the global minimum. Also, to ensure the convergence of the network and its clinical utility with useful results, the minimization is achieved in such a way that after a prespecified period of time, the energy function can reach a local minimum, close to the global minimum, and remains there ever after. We present here, segmentation data results for a subject diagnosed with a metastaric tumor in the brain