Clinical applications and recent advances in multimodal image fusion for hepatocellular carcinoma diagnosis: a narrative review.
Authors
Affiliations (4)
Affiliations (4)
- Department of Nephrology, The First Hospital of Anhui University of Science and Technology, Huainan, China.
- School of Computer Science and Engineering, Anhui University of Science & Technology, Huainan, China.
- School of Medicine, Anhui University of Science & Technology, Huainan, China.
- Anhui Provincial Academician Workstation of Anhui University of Science & Technology for Autoimmune Disease Diagnostic Technology, Huainan, China.
Abstract
Hepatocellular carcinoma (HCC) remains a major cause of cancer-related mortality worldwide, and patient outcomes depend strongly on early detection, accurate staging, and timely treatment planning. However, single-modality imaging cannot fully characterize tumor heterogeneity because each modality provides only partial anatomical, functional, or metabolic information. Multimodal image fusion (MIF) addresses this limitation by integrating complementary data from ultrasound (US), computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). This narrative review summarizes advances in MIF and its clinical applications in HCC. A targeted literature search of relevant databases was conducted, with a focus on recent and key English-language publications. The search strategy included terms related to "hepatocellular carcinoma", "CT", "MRI", "PET", "image registration", "image-guided ablation", and "artificial intelligence". This review summarizes commonly used image-fusion strategies and recent advances in their clinical applications for HCC, including lesion detection, staging, ablation guidance, and post-treatment assessment. It also discusses the emerging role of artificial intelligence (AI) in image registration, fusion, and interpretation. MIF has shown clear clinical promise in supporting precision diagnosis and personalized management of HCC. Although MIF has shown clear clinical promise, its broader adoption remains limited by technical interoperability, cost, workflow complexity, and the need for standardized training. Future progress in system integration, algorithm robustness, and multidisciplinary collaboration may further expand its value in precision diagnosis and personalized management of HCC.