Medical Image Fusion Based on Non-Subsampled Shearlet Transform and Spiking Cortical Model

Medical image fusion plays an important role in surgical guidance, disease diagnosis and treatment. The existing image fusion algorithms cannot preserve image details well or introduce unwanted artefacts. To address this problem, this paper proposes a novel medical image fusion method that combines the non-subsampled shearlet transform (NSST) with the spiking cortical model (SCM). This method firstly utilizes the NSST to decompose the registered source images to produce multiscale and multidirectional sub -bands, and processes these sub bands using the SCM to generate the corresponding firing mapping images. Then fused sub -band in each scale and direction will be obtained by selecting coefficients from the sub -bands according to local energy (LE) of the corresponding firing mapping images. Finally, the fused image is produced by applying the inverse NSST to the fused sub -bands. Simulation experiments have been made to compare the effects of our method with several state-of-the-art fusion methods. The performance of these methods is appreciated by human vision and objective evaluation metrics. The experimental results demonstrate that compared with other methods, our method performs better in detail preservation and artefact avoidance.