NVIDIA and C-BRAIN deploy open-source AI tools to speed dementia research

There are 57 million people living with dementia worldwide, and more than 10 million new cases are diagnosed each year. In a blog post published October 27, NVIDIA argues that the obstacle for researchers is not a lack of effort but too much data, too few experts to synthesize it, and a pace of discovery that cannot keep up with the disease.

The Consortium for Biomedical Research and AI in Neurodegeneration, known as C-BRAIN, is working with NVIDIA to change that. The consortium, led by Dr. Randall Bateman of Washington University School of Medicine in St. Louis, launched in November 2025 and brings together 18 member institutions across academia, philanthropy, pharma and research organizations.

Alzheimer's research has been constrained by two problems: too few high-impact findings, and data trapped in institutional silos. C-BRAIN calls the unpublished findings, negative results and governed proprietary data that have never been accessible to AI tools or human researchers "dark data."

Using that material without compromising patient privacy or institutional governance is hard. C-BRAIN runs a hybrid centralized and federated approach, in which models learn from data at each institution without the data ever leaving its walls. That kind of infrastructure is technically demanding to build and scale, which is part of what NVIDIA contributes.

On the hardware side, C-BRAIN currently runs 64 NVIDIA data center GPUs on premises alongside cloud nodes on Microsoft Azure, with additional capacity including Blackwell systems planned as the consortium scales. On the software side, it is incorporating NVIDIA's BioNeMo open developer platform and the MONAI framework into its research pipelines.

The consortium has already built two tools: OpenScientist, developed 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 co-scientist that takes a research question and dataset, then semi-autonomously generates hypotheses, runs analyses and produces verifiable, auditable results. InsightEngine tests hypotheses against the scientific literature, analyses user-supplied datasets, explores a biomedical knowledge graph and checks 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. Across C-BRAIN's current reach of more than 200 research groups, that could free up more than one million hours of research capacity annually at scale.

Looking further ahead, C-BRAIN aims to build a multimodal biological foundation model trained on neuroimaging, multi-omics, fluid biomarkers and longitudinal clinical measures. By year three, it hopes to identify new causal mechanisms of brain aging and Alzheimer's disease. "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," Bateman said.