High-Quality MRI and PET/SPECT Image Fusion Based on Local Laplacian Pyramid (LLP) and Adaptive Cloud Model (ACM) for Medical Diagnostic Applications

Abstract

Image fusion plays a major role in biomedical applications such as tumor detection, medical diagnostics, disease identification, etc. Generally, medical imaging modalities such as Positron Emission Tomography (PET)/Single Photon Emission Computed Tomography (SPECT) and Magnetic Resonance Images (MRI) are used to perform the fusion process for post-surgery analysis. MRI images generally have a single channel i.e. gray information about the skull.MRI images give anatomical data of soft tissues whereas PET/SPECT images give functional images of tissues. Therefore, combining MRI and PET/SPECT images give both structural as well as functional information. In this paper, a new approach for PET/SPECT and MRI image fusion using the Adaptive Cloud Model (ACM) based Local Laplacian Pyramid (LLP) is proposed to obtain the high quality fused output. To increase the sensitivity of the fusion, the RGB image is converted into Hue Intensity Saturation (HIS) color transform. LLP is applied to the gray level component of the MRI image and Intensity component of PET/SPECT images respectively. The Adaptive Cloud Model is used to perform the fusion of LLP coefficients. The inverse of LLP and HIS transform is applied to get the fused image in color domain. Performance evaluation shows that the proposed method gives better performance when compared to conventional techniques.