AI 如何改变数百万痴呆症患者的发现速度

如今全球有 5700 万人患有痴呆症。对于致力于帮助他们研究人员来说,障碍并非努力不足——而是数据太多、能综合这些数据的专家太少,以及发现速度跟不上疾病的发展。
神经退行性疾病生物医学研究与人工智能联盟(C-BRAIN)正与 NVIDIA 合作,利用 AI 改变这一局面。
阿尔茨海默病是当今全球最主要的健康挑战之一,每年新诊断的痴呆症病例超过 1000 万。尽管过去 15 年美国国立卫生研究院(NIH)的资助规模扩大到原来的 7 倍——达到每年 39 亿美元——但仍未找到能够阻止、预防或逆转该疾病的治疗方法。
瓶颈在于高影响力的科学发现不足、太多数据困在机构孤岛中,以及综合速度过慢,跟不上已发表的研究。
C-BRAIN 由 Randall Bateman 博士领导,由 18 家成员机构组成,涵盖学术、慈善、制药和研究组织,于 2025 年 11 月启动,旨在通过加速使新疗法成为可能的研究,帮助科学家实现更好、更快的发现。
“人工智能为我们打开了一个机会窗口,让我们能够以根本上不同于以往的方式开展研究,从而加速发现以帮助患者,”C-BRAIN 创始人兼主任、圣路易斯华盛顿大学医学院 Charles F. and Joanne Knight 神经病学杰出教授 Bateman 表示。他将他们的工作描述为“由科学家构建、为科学家服务的 AI 工具”。
该联盟正在构建一个系统,利用此前未被挖掘的阿尔茨海默病研究数据来支持科学家的工作。
阿尔茨海默病研究中至关重要的资产,是那些从未向 AI 工具或人类研究人员公开开放过的数据集。这包括未发表的研究发现、阴性结果,以及来自成员机构的受管制的专有数据。
“大量生命科学数据未能进入前沿模型,”C-BRAIN 首席技术官 Ade Ojewole 博士表示。“其后果就是,我们让大量智能白白闲置了。”
C-BRAIN 将此称为“暗数据”,它拖慢了阿尔茨海默病研究的进展。
部分暗数据被直接存入安全的计算机系统,供 AI 工具访问。为在不损害患者隐私或机构治理的前提下使用这些数据,C-BRAIN 采用了集中式与联邦式相结合的混合方法——AI 模型在各机构本地从数据中学习,而数据始终无需离开机构内部。这类基础设施的构建与扩展在技术上要求极高,而这正是 NVIDIA 为这一合作带来的部分能力。
C-BRAIN 目前在本地运行 64 块 NVIDIA 数据中心 GPU,同时在 Microsoft Azure 上运行基于云的节点;随着该联盟的雄心不断扩展,更多 NVIDIA GPU 算力(包括 NVIDIA Blackwell 系统)也将陆续到位。
为探索 AI 与加速计算如何在应对其技术挑战的同时推进阿尔茨海默病研究,C-BRAIN 正与 NVIDIA 合作,将 NVIDIA BioNeMo 开放开发者平台和 MONAI 框架等技术纳入 C-BRAIN 的研究流程。
“C-BRAIN 与 NVIDIA 的合作旨在释放这些模型的潜力,使其能够带来极具影响力和重要意义的东西,”Bateman 表示。
“随着我们不断拓展当今空间组学机器学习架构的建模边界,将需要大量的优化,”Ojewole 表示。“NVIDIA 的软件肯定会在其中发挥作用。”
C-BRAIN 的研究人员已经构建了两款工具:OpenScientist——在由研究科学家 Justin Reese 牵头的一项工作中与 Lawrence Berkeley National Laboratory 合作构建——以及与 387 Labs 合作构建的 InsightEngine。
OpenScientist 是一款开源的智能体式 AI 共同科学家(co-scientist),它接收科学家提出的研究问题和数据集,然后半自主地生成假设、运行分析,并产出可验证、可审计的洞见。InsightEngine 是一个开源研究平台,利用智能体式 AI 对照科学文献检验假设,分析用户提供的数据集,探索生物医学知识图谱,并对每一项论断进行证据核查,从而在研究人员投入成本高昂的实验之前捕捉幻觉。
在已发表的案例研究中,OpenScientist 在几分钟内完成了领域科学家估计需要数周到数月才能完成的分析,每次运行的成本低于 10 美元。以 C-BRAIN 目前覆盖 200 多个研究组的规模,这一能力可释放大量研究产能,在规模化后每年可能超过 100 万小时。
展望未来,C-BRAIN 的目标是构建一个多模态生物学基础模型,以其全部数据广度进行训练——神经影像、多组学、体液生物标志物和纵向临床测量指标。
到第三年,其目标是发现脑衰老和阿尔茨海默病的新因果机制,进而利用所有可用的研究和数据,迈向可检验的科学原理。
Bateman 表示:“能够思考和推理的 AI,有望加速各类发现,这些发现将带来强大且可落地的预防与治疗手段,惠及全体人群。”
译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。
查看英文原文(How AI Is Changing the Pace of Discovery for Millions Living With Dementia)
There are 57 million people worldwide living with dementia today. For the researchers working to help them, the obstacle isn’t a lack of effort — it’s too much data, too few experts to synthesize it and a pace of discovery that can’t keep up with the disease.
The Consortium for Biomedical Research and AI in Neurodegeneration (C-BRAIN), working with NVIDIA, is using AI to change that.
Alzheimer’s disease is one of today’s leading global health challenges, with over 10 million new dementia cases diagnosed each year. Despite a 7x increase in National Institutes of Health funding in the last 15 years — reaching $3.9 billion annually — no treatment has been found to stop, prevent or reverse the disease yet.
The bottlenecks are not enough high impact scientific findings, too much data living in institutional silos and too slow a pace of synthesis to keep up with the research that is published.
C-BRAIN, led by Dr. Randall Bateman and a consortium of 18 member institutions across academic, philanthropic, pharma and research organizations, launched in November 2025 to help scientists make better, faster discoveries by accelerating the research that makes new treatments possible.
“We have a window of opportunity with artificial intelligence to do research fundamentally different than we’ve ever done before, in a way that accelerates discoveries to help patients,” said Bateman, founder and director of C-BRAIN and the Charles F. and Joanne Knight Distinguished Professor of Neurology at the Washington University School of Medicine in St. Louis. He described their work as “AI tools built by scientists, for scientists.”
The consortium is building a system that harnesses previously untapped Alzheimer’s research data to support scientists’ work.
Speeding Research From Weeks to Hours
Essential assets in Alzheimer’s research are datasets that have never been publicly accessible to AI tools or human researchers. This includes unpublished findings, negative results and governed proprietary data from member institutions.
“A lot of life sciences data doesn’t make its way into the frontier models,” said Dr. Ade Ojewole, chief technology officer of C-BRAIN. “The consequence of that is we’re leaving a lot of intelligence on the table.”
C-BRAIN refers to this as “dark data,” which slows down Alzheimer’s research.
Some dark data is directly deposited into secure computer systems for AI tools to access. To use this data without compromising patient privacy or institutional governance, C-BRAIN uses a hybrid centralized and federated approach — where AI models learn from data at each institution without that data ever having to leave its walls. This kind of infrastructure is technically demanding to build and scale, which is part of what NVIDIA brings to the collaboration.
C-BRAIN currently runs 64 NVIDIA data center GPUs on premises alongside cloud-based nodes on Microsoft Azure, with additional NVIDIA GPU capacity, including NVIDIA Blackwell systems, on the horizon as the consortium’s ambitions scale.
To explore how AI and accelerated computing can advance Alzheimer’s research while navigating its technical demands, C-BRAIN is working with NVIDIA to incorporate technologies including the NVIDIA BioNeMo open developer platform and the MONAI framework for C-BRAIN’s research pipelines.
“The C-BRAIN collaboration with NVIDIA aims to unlock those models in a way that they can deliver something quite impactful and important,” said Bateman.
“A great deal of optimization is going to be needed as we push the boundaries of what today’s spatial omics machine learning architectures can model,” Ojewole said. “NVIDIA software is definitely going to come into play there.”
Researchers at C-BRAIN have already built two tools: OpenScientist — built with Lawrence Berkeley National Laboratory in an effort led by research scientist Justin Reese — and InsightEngine, built with 387 Labs.
OpenScientist is an open source agentic AI co-scientist that takes a scientist’s research question and dataset, then semi-autonomously generates hypotheses, runs analyses and produces verifiable, auditable insights. InsightEngine is an open source research platform that uses agentic AI to test hypotheses against the scientific literature, analyzing user-supplied datasets, exploring a biomedical knowledge graph and checking every claim against evidence to catch hallucinations before researchers commit to costly experiments.
In published case studies, OpenScientist completed analyses in minutes that domain scientists estimated would take weeks to months, at a cost of under $10 per run. At C-BRAIN’s current reach of over 200 research groups, this capability could release substantial research capacity, potentially exceeding one million hours annually at scale.
From Correlation to Cause
Looking ahead, C-BRAIN aims to build a multimodal biological foundation model trained on the full breadth of its data — neuroimaging, multi-omics, fluid biomarkers and longitudinal clinical measures.
By year three, it aims to discover new causal mechanisms of brain aging and Alzheimer’s disease, moving to testable scientific principles using the full scope of available research and data.
Bateman said, “AI that can think and reason has the potential to accelerate discoveries that will lead to powerful and implementable preventions and treatments for the entire population.”
正文保留英文原文,标题与摘要由机器翻译。
原文发布于 NVIDIA Blog(How AI Is Changing the Pace of Discovery for Millions Living With Dementia),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。