arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2506.02470cs.AI
收藏

一种具备强大LLM推理能力的智能多模态医疗保健副驾驶

A Smart Multimodal Healthcare Copilot with Powerful LLM Reasoning

Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly (LILY), NTU(联合NTU-UBC老龄化积极生活卓越研究中心(LILY),NTU);College of Computing and Data Science, Nanyang Technological University (NTU), Singapore(计算与数据科学学院,南洋理工大学(NTU),新加坡);Tan Tock Seng Hospital, Singapore(坦 tok sing 医院,新加坡);Woodlands Health, Singapore(伍德兰兹健康,新加坡)

Xuejiao Zhao, Siyan Liu, Su-Yin Yang, Chunyan Miao

AI总结:

MedRAG是一种多模态医疗保健副驾驶,通过知识图谱引导的检索增强生成和LLM推理整合多源诊断信息,降低误诊风险,并在公共与私有数据集上优于现有模型。

摘要:

误诊对全球医疗系统造成重大危害,导致成本上升和患者风险增加。MedRAG是一种智能多模态医疗保健副驾驶,配备强大的大语言模型(LLM)推理能力,旨在增强医疗决策。它支持多种输入模态,包括非侵入式语音监测、一般医疗咨询和电子健康记录。MedRAG在诊断、治疗、用药和后续追问方面提供建议。利用由知识图谱引导推理增强的检索增强生成,MedRAG检索并整合关键诊断信息,降低误诊风险。该模型已在公共和私有数据集上进行了评估,性能优于现有模型,并提供更具体、更准确的医疗保健辅助。MedRAG的演示视频可在https://www.youtube.com/watch?v=PNIBDMYRfDM查看,源代码可在https://github.com/SNOWTEAM2023/MedRAG获取。

英文摘要:

Misdiagnosis causes significant harm to healthcare systems worldwide, leading to increased costs and patient risks. MedRAG is a smart multimodal healthcare copilot equipped with powerful large language model (LLM) reasoning, designed to enhance medical decision-making. It supports multiple input modalities, including non-intrusive voice monitoring, general medical queries, and electronic health records. MedRAG provides recommendations on diagnosis, treatment, medication, and follow-up questioning. Leveraging retrieval-augmented generation enhanced by knowledge graph-elicited reasoning, MedRAG retrieves and integrates critical diagnostic insights, reducing the risk of misdiagnosis. It has been evaluated on both public and private datasets, outperforming existing models and offering more specific and accurate healthcare assistance. A demonstration video of MedRAG is available at: https://www.youtube.com/watch?v=PNIBDMYRfDM. The source code is available at: https://github.com/SNOWTEAM2023/MedRAG.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑