Multimodal medical image fusion using empirical wavelet decomposition and local energy maxima

In the clinical environment, medical imaging plays a prominent role in assisting radiologists/ doctors. The information present in the images is crucial especially during the time of diagnosis. This can be considerably improved with the help of multimodal medical image fusion approach, by integrating the information from various imaging modalities. In recent years, various methodologies are developed to fuse medical images. However, the multimodal medical image fusion is still a challenging issue, due to the degradation of medical images at the acquisition phase. To address this problem, we combine the complementary information from the various distinct imaging modalities such as MRI, PET, and SPECT by reducing the distortion using empirical wavelet transform (EWT) representation and local energy maxima (LEM) fusion rule. In EWT, the basis functions are selected optimally, depending on the nature of the input image. It results in preserving critical information like edges which are crucial in image fusion. The use of LEM highlights, vital information with the help of local energy constraint. We compared the proposed methodology with the state -of- the -art approaches and observed improved performance in terms of visual quality as well as fusion metrics.