Applications and Prospects of General Large Language Models in Medical Imaging

Medical imaging plays a central role in modern medicine, underpinning disease diagnosis, staging, therapeutic evaluation, and longitudinal follow-up. However, conventional image interpretation remains limited by low efficiency, variable consistency, and heavy reliance on clinical expertise. Recent advances in artificial intelligence (AI)—particularly deep learning, radiomics, and general large language models (LLMs)—have opened new avenues for intelligent image analysis and clinical decision-making. Leveraging multimodal data integration and natural language generation, LLMs enable comprehensive interpretation that unifies imaging findings with clinical records, laboratory tests, and molecular profiles, thereby improving the efficiency, consistency, and interpretability of diagnostic and prognostic workflows^{[<xref ref-type="bibr" rid="R1">1</xref>]}. Here, we provide a systematic overview of the evolution and applications of AI in medical imaging, examine key challenges in data quality, algorithmic development, and clinical translation, and highlight future directions including multimodal integration, human–AI collaboration, and privacy-preserving frameworks. Collectively, emerging evidence indicates that LLMs and allied AI technologies are reshaping medical imaging toward more intelligent, individualized, and full-cycle clinical support, offering critical momentum for precision medicine.