Hybrid method for multimodality medical image fusion using Discrete Wavelet Transform and Entropy concepts with quantitative analysis

Medical image fusion is the process of deriving imperative information from multimodality medical images. This derived information can be used for various purposes like, diagnosing diseases, detecting the tumor, surgery treatment and so on. This type of information cannot be obtained using single modality image. Therefore, the drawbacks of single modality medical image has paved the way for the process of combining different modality images such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and Single Photon Emission Computed Tomography (SPECT) into a single image. Hence, the above described procedure of combining multimodality images in to a single fused image can be done using image fusion techniques. This paper proposes a new image fusion technique namely; Discrete Wavelet Transform-Averaging-Entropy-Principle Component Analysis method [DWT-A-EN-PCA] and the results of proposed system are compared with other existing fusion techniques using quantitative metrics such as, Entropy (EN), Signal to Noise Ratio (SNR) and Fusion Symmetric (FS) for performance evaluation. From the experimental result it is observed that Entropy (EN) provides better quality of information for proposed method, Signal to Noise Ratio (SNR) shows less noise ratio for proposed method, and Fusion Symmetric (FS) shows very less information loss during the fusion of proposed method.