发表机构
Median Technologies; The University of Osaka; Université de Paris Cité; AP-HP; Hôpital Universitaire Necker Enfants Malades; Memorial Sloan Kettering Cancer Center; Massachusetts General Hospital; Sorbonne Université（Median Technologies; 大阪大学; 巴黎西岱大学; 巴黎公立医院集团; 内克尔儿童医院; 纪念斯隆凯特琳癌症中心; 马萨诸塞总医院; 索邦大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究评估基础模型MedGemma在Lung-RADS筛查中的性能，发现微调后AUC达0.83，虽低于放射科医生平均0.90，但模型输出确定性高，可作为临床决策支持的互补工具。
AI 中文摘要
基础模型最近在广泛的医学影像任务中展示了强大的能力。然而，它们在结构化临床解读场景中的性能仍未得到充分探索。在肺癌筛查中，尽管存在如Lung-RADS等标准化框架，解读变异性仍然存在。在本研究中，我们评估了MedGemma，一个源自Gemini的医学通用基础模型及其针对肺癌检测与诊断进行微调的版本，并与放射科医生在NLST数据集上执行Lung-RADS v2022评估进行比较。十二名放射科医生在多读者设计中独立评估每个病例，从而能够量化读者间变异性。放射科医生实现了平均AUC为0.90，读者间存在显著变异性（范围：0.80-0.94）。原生基础模型实现了AUC为0.70，未能达到临床相关性能。相比之下，微调显著提升了性能至AUC为0.83，使模型处于个体放射科医生性能的较低范围内。这些发现凸显了峰值准确性与预测一致性之间的权衡。与放射科医生不同，在固定条件下，模型产生确定性输出，消除了相同输入下的运行间变异性，这与放射科医生中观察到的读者间变异性形成对比。这支持了微调基础模型作为临床决策支持潜在互补工具的作用，特别是在专业知识有限的场景中。然而，评估是在来自NLST的病例富集队列上进行的，未考虑现实世界患病率或外部验证，限制了直接的临床泛化。
英文摘要
Foundation models have recently demonstrated strong capabilities across a wide range of medical imaging tasks. However, their performance in structured clinical interpretation settings remains insufficiently explored. In lung cancer screening, interpretative variability persists despite standardized frameworks such as Lung-RADS. In this study, we evaluate MedGemma, a medical general-purpose foundation model derived from Gemini and its fine-tuned version adapted for lung cancer detection and diagnosis, compared against radiologists performing Lung-RADS v2022 assessment on the NLST dataset. Twelve radiologists independently evaluated each case in a multi-reader design, enabling quantification of inter-reader variability. Radiologists achieved a mean AUC of 0.90, with substantial variability across readers (range: 0.80-0.94). The native foundation model achieved an AUC of 0.70, failing to reach clinically relevant performance. In contrast, fine-tuning significantly improved performance to an AUC of 0.83, placing the model within the lower range of individual radiologists performance. These findings highlight a trade-off between peak accuracy and prediction consistency. Unlike radiologists, under fixed conditions, the model produces deterministic outputs, removing inter-run variability under identical inputs, in contrast to inter-reader variability observed among radiologists. This supports the role of fine-tuned foundation models potential complementary tools for clinical decision support, particularly in settings with limited expertise. However, evaluation is performed on a case-enriched cohort from NLST and does not account for real-world prevalence or external validation, limiting direct clinical generalization.
CommentsMICCAI CAPTION 2026
