Abstract
This paper is concerned with the topic of medical image fusion. Bothe Magnetic Resonance (MR) and Computed Tomography (CT) images are fused using a Convolutional Neural Network (CNN). The CNN is composed of several layers that comprise convolutional layers, pooling layers and a fully-connected layer. The proposed approach comprises a hierarchy that contains focus detection, initial segmentation, consistency verification and fusion. The objective of the utilization of the CNN is to generate a focus map from the two input images. The focus map is hanged to a binary map. Some sort of image post-processing is used to remove noise and undesired small objects. Simulation results of the fusion of MR and CT images reveal images with high visual quality with much details. This can help in the utilization of these images for further diagnosis applications.