自动食道癌症分期从自由文本放射学报告:大型语言模型评估研究
Yao Yao1, Xingxing Cen1, Lu Gan2
1Information Center, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, No 241, West Huaihai Road, Shanghai, 200030, China, 86 22200000.
JMIR medical informatics
|October 17, 2025
概括
大型语言模型 (LLM) 在食道癌症分期方面表现有前途. 具有可解释推理的INF-72B LLM与临床医生相比,在从放射学报告中分期癌症方面取得了更高的准确性.
科学领域:
- 在瘤学瘤学.
- 医疗成像医学成像
- 人工智能的人工智能
背景情况:
- 精确的食道癌分期对于患者的预后和治疗计划至关重要.
- 临床医生手动解释放射学报告在准确性和一致性方面存在挑战.
- 大型语言模型 (LLM) 在食道癌分期中的应用是一个新兴的研究领域.
研究的目的:
- 评估和比较三个本地部署的LLM与临床医生在术前食道癌症分期中的表现.
- 评估不同提示策略的实用性,包括可解释推理 (IR) 方法,用于基于LLM的癌症分期.
主要方法:
- 上海胸部医院200名食道癌患者的回顾性研究.
- 分析了1134份中国自由文本放射学报告,以术后病理阶段作为参考标准.
- 三个LLM (INF-72B,Qwen2.5-72B,LLaMA3.1-70B) 用零射击,思维链和IR提示策略来确定瘤,结节和整体阶段.
主要成果:
- 与临床医生相比,INF-72B与IR策略相结合,实现了明显更高的整体分期精度 (61.5%) 和F1得分 (0.60),精度为39.5%,F1得分为0.39).
- 具有IR的Qwen2.5-72B也超过了临床医生,显示了46%的准确性和0.51的F1得分.
- 与临床医生相比,LLaMA3.1-70B在分期表现上没有显示出统计学上显著的差异.
结论:
- 从放射学报告来看,LLM,特别是INF-72B和拟议的IR策略,可以准确地确定食道癌的阶段.
- 这种由人工智能驱动的方法提供了一个透明和可验证的推理过程,提高了诊断可靠性.
- LLM显示出作为有价值的决策支持工具的潜力,以增强食道癌症分期的临床专业知识.
更多相关视频
03:14Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
1.0K
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
485
相关概念视频
02:43Mouse Models of Cancer Study
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
02:43Mouse Models of Cancer Study
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...