A Comprehensive Framework for the Fusion of Anatomical (CT, MRI) and Functional (PET, SPECT) Medical Images for Enhanced Diagnostic Precision

Abstract

Medical image fusion aims to integrate complementary information from multimodal imaging techniques to provide a more comprehensive visualization for diagnosis and treatment planning. This study presents a novel framework for fusing anatomical images (CT and MRI) with functional images (PET and SPECT), addressing challenges related to resolution disparity, spatial misalignment, and modality-specific noise. The proposed method utilizes a hybrid approach combining deep learning-based registration, wavelet-based fusion, and contrast enhancement techniques. Evaluation on clinical datasets demonstrates improved anatomical-functional correspondence, enhanced tumor delineation, and superior performance on quantitative metrics such as mutual information (MI), structural similarity index (SSIM), and edge preservation index. The results support the clinical utility of fused images in oncology, neurology, and cardiology.