摘要：

通用医学大语言模型(LLM)在脑肿瘤领域存在专业数据匮乏、临床适应性不足及生成内容准确性有限等问题, 提出一种专用于脑肿瘤诊疗领域的大语言模型BrainTumorLLM。该模型基于Meta-LLaMA-3-8B-Instruct模型, 通过监督微调(SFT)和人类反馈强化学习(RLHF)技术优化, 结合自建的高质量脑肿瘤问答数据集BrainTumorQA进行训练。数据集采用宏观-微观协同的构建框架, 共包含11 000条问答对, 涵盖宏观医学知识(症状、诊断方法、治疗方案)及微观临床病例, 并通过脱敏处理与信息约束策略保障数据安全。在技术实现中, 采用低秩适配(LoRA)技术提升训练效率, 设计宏观与微观两级提示模板, 引导模型生成专业化回答, 并引入RLHF, 通过专家偏好驱动优化机制以及近端策略优化(PPO)算法强化生成内容的临床一致性。实验结果表明, BrainTumorLLM在脑肿瘤问答任务中显著优于通用及医学领域模型, 在自动评估环节, 其BLEU-1、BLEU-2分别达到了0.338 3和0.268 4, ROUGE-1、ROUGE-2和ROUGE-L得分分别为0.323 7、0.146 6和0.261 1, 与基准模型相比困惑度从20.362降至7.674, 充分显示了所提模型在脑肿瘤诊疗领域的专业性、精准性及临床应用潜力, 为脑肿瘤的诊断、治疗决策以及医学科研等工作提供有力的智能化辅助支持。

关键词: 大语言模型, 脑肿瘤问答, 监督微调, 人类反馈强化学习, 临床决策支持

Abstract:

To address the challenges faced by general-purpose medical Large Language Model (LLM) in the field of brain tumor care—namely the scarcity of domain-specific data, limited clinical adaptability, and insufficient accuracy of generated content. This paper proposes BrainTumorLLM, a specialized LLM tailored for brain tumor diagnosis and treatment. Built upon the Meta-LLaMA-3-8B-Instruct foundation model, BrainTumorLLM is optimized via Supervised Fine-Tuning (SFT) and Reinforcement Learning with Human Feedback (RLHF) and trained using a self-constructed, high-quality dataset named BrainTumorQA. This dataset comprises 11 000 question-answer pairs, encompassing both macro-level medical knowledge (symptoms, diagnostic methods, and treatment strategies) and micro-level clinical cases, with privacy safeguarded via anonymization and information constraint strategies. From a technical perspective, Low-Rank Adaptation (LoRA) is employed to enhance the training efficiency. A two-tier prompting framework is designed to guide the model in generating domain-specific responses at both the macro and micro levels. Furthermore, RLHF is integrated using an expert preference-driven optimization mechanism and a Proximal Policy Optimization (PPO) algorithm, reinforcing the clinical consistency of the generated content. The experimental results demonstrate that BrainTumorLLM significantly outperforms both general-purpose and medical-domain models in brain tumor-related question-answering tasks. In automatic evaluations, it achieves BLEU-1 and BLEU-2 scores of 0.338 3 and 0.268 4, respectively, and ROUGE-1, ROUGE-2, and ROUGE-L scores of 0.323 7, 0.146 6, and 0.261 1, respectively. Moreover, the perplexity of the model is substantially reduced from 20.362 (base model) to 7.674, highlighting its domain-specific precision, professional accuracy, and potential for clinical applications. BrainTumorLLM is a robust AI-powered tool that supports brain tumor diagnosis, treatment planning, and medical research.

Key words: Large Language Model(LLM), brain tumor question-answering, Supervised Fine-Tuning(SFT), Reinforcement Learning with Human Feedback(RLHF), clinical decision support