An Agentic Generative Large Language Model for Treatment Planning of Colorectal Cancer
作者: Mengxian Lyu, Cheng Peng, Tim Jang, Ang Li, Mengyuan Zhang, Ziyi Chen, Leighton Elliott, Tianshi Liu, Lidice Galindo, Chiranjeevi Sainatham, Oscar F. Borja-Montes, Kaleb E. Smith, Ying Zhang, Lichao Sun, Jiang Bian, Gloria Lipori, Duane A. Mitchell, Elizabeth A. Shenkman, Yi Guo, Thomas J. George, Yonghui Wu
分类: cs.CL
发布日期: 2026-08-10
💡 一句话要点
提出GatorOnco以解决结直肠癌治疗规划中的复杂推理问题
🎯 匹配领域: 支柱二:RL算法与架构 (RL & Architecture) 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 结直肠癌 治疗规划 大型语言模型 代理推理 临床指南 生物医学文本 强化学习 领域适应
📋 核心要点
- 现有大型语言模型在高风险治疗规划中面临复杂推理和安全性问题,限制了其应用。
- GatorOnco通过代理检索增强生成方法,动态整合临床指南,提升治疗规划的准确性和安全性。
- 在临床评估中,GatorOnco在可读性和完整性方面显著优于开源模型,表现出专家级的效果。
📝 摘要(中文)
精准肿瘤学中的治疗规划需要综合异构患者信息与快速发展的临床指南,以确保符合指南的护理。尽管大型语言模型在许多诊断任务中显示出潜力,但由于复杂推理、遵循及时临床指南和安全性问题,其在高风险治疗规划中的应用受到限制。本研究提出了GatorOnco,一个用于结直肠癌治疗规划的代理大型语言模型。GatorOnco基于2820亿个生物医学文本标记开发,包括来自UF Health的1660亿个临床文本。我们实施了一种领域适应方法,结合了预训练、模型合并、两阶段后训练方法和基于代理的强化学习。代理检索增强生成(RAG)方法动态整合了时间敏感的临床指南。在由五位UF Health肿瘤科医生进行的盲随机临床评估中,GatorOnco显著优于开源大型语言模型,并达到了与UF Health肿瘤科医生相当的专家级表现。
🔬 方法详解
问题定义:本研究旨在解决结直肠癌治疗规划中的复杂推理问题,现有方法在遵循临床指南和安全性方面存在不足。
核心思路:GatorOnco通过代理检索增强生成(RAG)方法,动态整合时间敏感的临床指南,以提高治疗规划的准确性和安全性。
技术框架:GatorOnco的整体架构包括预训练、模型合并、两阶段后训练和基于代理的强化学习四个主要模块,确保模型在临床应用中的有效性。
关键创新:本研究的关键创新在于将代理推理与大规模领域适应相结合,显著提升了生成AI在高风险癌症治疗规划中的应用能力。
关键设计:在模型训练中,采用了2820亿个生物医学文本标记,特别是1660亿个来自UF Health的临床文本,确保了模型的知识基础和适应性。
🖼️ 关键图片
📊 实验亮点
在盲随机临床评估中,GatorOnco显著优于开源大型语言模型(P < 0.01),在可读性(4.46 vs. 4.19, P < 0.01)和完整性(3.91 vs. 3.52, P < 0.01)方面获得更高评分,同时在正确性、时效性和安全性方面表现与专家相当。
🎯 应用场景
GatorOnco的研究成果在精准肿瘤学领域具有广泛的应用潜力,能够辅助医生进行更为准确和安全的治疗规划。未来,该模型可能在其他癌症类型的治疗规划中推广应用,提升整体医疗决策的智能化水平。
📄 摘要(原文)
Treatment planning in precision oncology requires synthesizing heterogeneous patient information with rapidly evolving clinical guidelines to ensure guideline-concordant care. While large language models (LLMs) show promise in many diagnostic tasks, their adoption for high-stakes treatment planning is hindered by complex reasoning, adherence to timely clinical guidelines, and safety concerns. In this study, we present GatorOnco, an agentic LLM for colorectal cancer (CRC) treatment planning. GatorOnco is developed using a total of 282 billion tokens of biomedical text, including healthcare system-scale clinical text comprising 166 billion tokens from UF Health. We implemented a domain-adaptation method that integrates pre-training, model merging, a two-stage post-training approach, and agent-based reinforcement learning. An agentic retrieval-augmented generation (RAG) approach dynamically integrates time-sensitive clinical guidelines into the reasoning process. In a blind, randomized clinical evaluation conducted by five UF Health oncologists, GatorOnco significantly outperformed open-source LLMs (P < 0.01) and achieved expert-level performance comparable to UF Health oncologists. Compared with expert oncologists, GatorOnco received significantly higher ratings for readability (4.46 vs. 4.19, P < 0.01) and completeness (3.91 vs. 3.52, P < 0.01), while showing statistically comparable performance in correctness (4.09 vs. 4.11, P = 0.921), currency (4.04 vs. 3.98, P = 0.478), and safety (4.22 vs. 4.22, P = 0.999). These findings demonstrate that integrating agentic reasoning with large-scale domain adaptation can help bridge the gap for generative AI in high-stakes cancer treatment planning.