Recent advances in imaging and artificial intelligence (AI) for quantitative assessment of multiple myeloma
Yongshun Liu,Wenpeng Huang,Yihan Yang,Weibo Cai,Zhaonan Sun
DOI: https://doi.org/10.62347/NLLV9295
2024-08-25
Abstract:Multiple myeloma (MM) is a malignant blood disease, but there have been significant improvements in the prognosis due to advancements in quantitative assessment and targeted therapy in recent years. The quantitative assessment of MM bone marrow infiltration and prognosis prediction is influenced by imaging and artificial intelligence (AI) quantitative parameters. At present, the primary imaging methods include computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). These methods are now crucial for diagnosing MM and evaluating myeloma cell infiltration, extramedullary disease, treatment effectiveness, and prognosis. Furthermore, the utilization of AI, specifically incorporating machine learning and radiomics, shows great potential in the field of diagnosing MM and distinguishing between MM and lytic metastases. This review discusses the advancements in imaging methods, including CT, MRI, and PET/CT, as well as AI for quantitatively assessing MM. We have summarized the key concepts, advantages, limitations, and diagnostic performance of each technology. Finally, we discussed the challenges related to clinical implementation and presented our views on advancing this field, with the aim of providing guidance for future research.