Evaluation of the MRI Images Matching Using Normalized Mutual Information Method and Preprocessing Techniques

One of the common methods for medical diagnosis is Magnetic Resonance Imaging (MRI), a safe, non-invasive method. During each imaging session a patient’s position may be different, therefore comparison of two sequences can become difficult. The primary goal of this work is preparation of an optimal algorithm for co-registration of T1 and T2 weighted MRI images. To adjust co-registration sensitivity, different preprocessing methods to perform normalizations and edge detection were used. The obtained results allow to increase quality of the co-registration process.