A New Robust Adaptive Fusion Method for Double-Modality Medical Image PET/CT

A new robust adaptive fusion method for double-modality medical image PET/CT is proposed according to the Piella framework. The algorithm consists of following three steps. Firstly, registered PET and CT images are decomposed using nonsubsampled contourlet transform (NSCT). Secondly, in order highlight lesions low-frequency image, components fused by pulse-coupled neural network (PCNN) that has a higher sensitivity featured area with low intensities. With regard high-frequency subbands, Gauss random matrix used compression measurements, histogram distance between every two corresponding subblocks high coefficient employed as match measure, regional energy activity measure. factor d then calculated measure measurement value factor, reconstructed orthogonal matching pursuit after fusion. Thirdly, final acquired through NSCT inverse transformation image. To validate algorithm, four comparative experiments were performed: experiment other algorithms, comparison different measures, results lung cancer (20 groups). experimental showed could better retain show lesion information, superior algorithms based on both subjective objective evaluations.