Multimodality image fusion in frequency domain for radiation therapy

Multimodality Medical image fusion is the process of extracting complementary information from various modality medical images and combing it for better visualization, accurate diagnosis and appropriate treatment planning. The combined single image is merging of anatomical and physiological variations. It allows accurate localization of cancer tissues and more helpful for estimation of target volume for radiation. The multimodal fusion algorithms presented in this paper utilizes Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT), Dual Tree Complex Wavelet Transform (DT-CWT), and Daubechies Complex Wavelet Transform (DCWT) to extract features of Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) images which are then combined based on various fusion rules. The performance and effectiveness of the algorithms are evaluated using Standard Deviation (σ_fus), Entropy (En), Fusion Factor (FusFac), Cross correlation (Rcorr) and Cross Entropy (CEn). The fused images of DCWT are superior over other frequency domain algorithms as per subjective and objective analysis.