arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.21597cs.AIcs.CV
收藏

Cerebra:一个多学科AI平台用于多模态痴呆症特征化和风险评估

Cerebra: A Multidisciplinary AI Board for Multimodal Dementia Characterization and Risk Assessment

Department of Biomedical Data Science, Stanford University(斯坦福大学生物医学数据科学系);Department of Radiation Oncology, Stanford University(斯坦福大学放射肿瘤科);Center for Data Science, New York University(纽约大学数据科学中心);J. Crayton Pruitt Family Department of Biomedical Engineering, University of Florida(佛罗里达大学杰·克雷顿·普瑞特家庭生物医学工程系);Department of Biomedical Engineering and Informatics, Indiana University Indianapolis(印第安纳大学印第安纳波利斯分校生物医学工程与信息学系);Department of Neurology, NYU Grossman School of Medicine(纽约大学格罗斯曼医学院神经科);Department of Neuroscience, NYU Grossman School of Medicine(纽约大学格罗斯曼医学院神经科学系);UF Health Family Medicine – Haile Plantation(佛罗里达大学健康家庭医学-海莱种植园);Department of Family Medicine, Indiana University School of Medicine(印第安纳大学医学院家庭医学系);Department of Biostatistics and Health Data Science, Indiana University School of Medicine(印第安纳大学医学院生物统计学与健康数据科学系);Scripps Research Translational Institute(斯克里普斯研究转化研究所);Courant Institute, New York University(纽约大学柯朗研究所);Center for Biomedical Informatics, Regenstrief Institute(再生斯蒂夫研究所生物医学信息中心);Center for Cognitive Aging and Memory, McKnight Brain Institute, University of Florida(佛罗里达大学麦金托神经研究所认知衰老与记忆中心);Population Health Department, NYU Langone Health(纽约大学兰戈恩健康人口健康部);Radiology Department, NYU Langone Health(纽约大学兰戈恩健康放射科)
Sheng Liu, Long Chen, Zeyun Zhao, Qinglin Gou, Qingyue Wei, Arjun Masurkar, Kevin M. Spiegler, Philip Kuball, Stefania C. Bray, Megan Bernath, Deanna R. Willis,… 展开作者

Sheng Liu, Long Chen, Zeyun Zhao, Qinglin Gou, Qingyue Wei, Arjun Masurkar, Kevin M. Spiegler, Philip Kuball, Stefania C. Bray, Megan Bernath, Deanna R. Willis, Jiang Bian, Lei Xing, Eric Topol, Kyunghyun Cho, Yu Huang, Ruogu Fang, Narges Razavian, James Zou

AI总结:

Cerebra通过多智能体协调处理电子病历、临床笔记和医学影像,提供可视化分析和对话界面,提升临床决策支持的可解释性和鲁棒性。

摘要:

现代临床实践日益依赖对异构、动态和不完整患者数据的推理。尽管多模态基础模型的进步提升了各种临床任务的性能,但大多数现有模型仍静态、不透明且与现实世界临床流程不匹配。我们提出了Cerebra,一个交互式多智能体AI团队,协调专门处理电子病历、临床笔记和医学影像分析的智能体。这些输出合成到面向临床医生的仪表盘中,结合可视化分析和对话界面,使医生能够在诊疗过程中检查预测并 contextualize 风险。Cerebra通过操作结构化表示支持隐私保护部署,并在模态不完整时保持稳健。我们使用覆盖300万患者数据的多机构大规模数据集评估了Cerebra。Cerebra在多模态任务中持续优于最先进的单模态模型和大型多模态语言模型基线。在痴呆风险预测中,其AUROCs达到0.80,比最强的单模态模型0.74和语言模型基线0.68更高。在痴呆诊断中,其AUROC为0.86,在生存预测中,C-index为0.81。在与经验医生的读者研究中,Cerebra显著提高了专家表现,前瞻性痴呆风险估计的准确率提高了17.5个百分点。这些结果展示了Cerebra在临床护理中可解释、稳健决策支持的潜力。

英文摘要:

Modern clinical practice increasingly depends on reasoning over heterogeneous, evolving, and incomplete patient data. Although recent advances in multimodal foundation models have improved performance on various clinical tasks, most existing models remain static, opaque, and poorly aligned with real-world clinical workflows. We present Cerebra, an interactive multi-agent AI team that coordinates specialized agents for EHR, clinical notes, and medical imaging analysis. These outputs are synthesized into a clinician-facing dashboard that combines visual analytics with a conversational interface, enabling clinicians to interrogate predictions and contextualize risk at the point of care. Cerebra supports privacy-preserving deployment by operating on structured representations and remains robust when modalities are incomplete. We evaluated Cerebra using a massive multi-institutional dataset spanning 3 million patients from four independent healthcare systems. Cerebra consistently outperformed both state-of-the-art single-modality models and large multimodal language model baselines. In dementia risk prediction, it achieved AUROCs up to 0.80, compared with 0.74 for the strongest single-modality model and 0.68 for language model baselines. For dementia diagnosis, it achieved an AUROC of 0.86, and for survival prediction, a C-index of 0.81. In a reader study with experienced physicians, Cerebra significantly improved expert performance, increasing accuracy by 17.5 percentage points in prospective dementia risk estimation. These results demonstrate Cerebra's potential for interpretable, robust decision support in clinical care.

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。

↑