Multimodal medical image fusion towards future research: A review

Abstract

Medical imaging has been widely used to diagnose various disorders over the past 20 years. Primary challenges in medicine include accurate disease identification and improved therapies. It is challenging for the medical experts to diagnose diseases using a single imaging modality. The fusion of two or more images obtained from different imaging modalities is known as multi modal image fusion (MMIF).The fused image contains complementary information for all the input images. The main objective of MMIF is to obtain complementary information (structural and spectral) from input images to improve the quality and clear assessment of medical related problems. The aim of fusion process is not only to reduced the amount of data but construct image having more useful and complementary information which are understandable for human and computer. This review provides a detailed overview of: (i) medical imaging modalities, (ii) multimodal medical image databases, (iii) MMIF steps/rules, (iv) MMIF methods, (v) modalities integration, (vi) performance evaluation and empirical results, (vii) current modalities strengths and limitations, and (viii) future directions. This review is expected to be useful in establishing a solid foundation for the development of more valuable medical image fusion methods for clinical diagnosis. This review presented the detailed studies on the multimodal databases, research trends in imaging modality grouping, and fusion steps which are the critical areas in MMIF. Furthermore, current challenges and future directions are thoroughly discussed. (c) 2023 The Authors. Published by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).