CT and MR Image Fusion based on Guided filtering and Phase congruency in Non-Subsampled Shearlet Transform domain

Abstract

Multi-modal image fusion plays a very important role in the early diagnosis and treatment of diseases. The different types of scans like MRI, PET, CT, and SPECT, etc. are used by medical practitioners to interpret the problems in the human body. The information present in these scans is very important at the time of treatment which can be enhanced by multisensory medical image fusion. Medical image fusion is the method that fuses images from different sources to give better results which helps in better understanding. The proposed method of medical image fusion improves the quality of human visual perception of an object or a scene by combining the fine details given by multiple sensor data. This is achieved by using a guided image filter (GIF) and Non-subsampled Shearlet transform (NSST) with integrated phase congruency based fusion rules. The GIF gives the fine details which are then fused using the above transform technique. This method is validated on CT and MRI images. The experimental results reveal the effectiveness of the proposed integrated activity measures with consistency verification fusion in terms of the image quality and quantitative assessment.