Adaptive PET/CT fusion based on Metabolic Activity Measurement

Cancer is one of the primary causes of morbidity and mortality in the developing countries. Recently, to infer complex decisions during diagnosis and treatment planning, multi-modality imaging plays an important role. So, accurate anatomic localization of functional abnormalities is desirable. This paper aims to merge PET image which holds the information about the glucose metabolism representing the activity of tissues and the CT image that has a detailed, high-resolution anatomy. The proposed method calculates the adaptive weight for each pixel based on the metabolic activity rate and performs the fusion operation. The proposed algorithm is compared with some of the spatial domain image fusion algorithms using the non-reference image quality, fusion and error metrics. It is found that the proposed adaptive-weighted algorithm excels in performance over the other methods.