News
C-BRAIN says AI tools are helping researchers analyze Alzheimer’s data faster, using secure infrastructure and open-source systems.
A dementia research consortium backed by NVIDIA says artificial intelligence is helping cut the time needed to move from data to insight, as scientists face a global disease burden that remains large and growing. In material published by NVIDIA on Oct. 27, 2026, the Consortium for Biomedical Research and AI in Neurodegeneration, or C-BRAIN, said its tools are designed to help researchers work through large volumes of Alzheimer’s data more quickly and securely.
Evidence and context
The consortium’s reporting says 57 million people worldwide are living with dementia and that Alzheimer’s disease accounts for a large share of cases. Background material provided with the source states that the World Health Organization’s dementia fact sheet, updated Oct. 26, 2026, cites an estimated 57 million people living with dementia worldwide in 2021, with nearly 10 million new cases each year and Alzheimer disease contributing to about 60% to 70% of cases. The WHO material also says dementia is the fifth-leading cause of death globally and a major cause of disability and dependency among older adults.
C-BRAIN, led by Dr. Randall Bateman and a consortium of 18 member institutions spanning academic, philanthropic, pharmaceutical and research organizations, launched in November 2025, according to the source. Bateman said the group sees artificial intelligence as a way to accelerate discovery, describing the effort as “AI tools built by scientists, for scientists.” The consortium says the problem is not only scientific effort but also the difficulty of synthesizing large, siloed datasets quickly enough to keep pace with publication volume and disease complexity.
The source says C-BRAIN is building a hybrid centralized and federated infrastructure so models can learn from data at each institution without the data leaving those sites. The consortium says it currently runs 64 NVIDIA data center GPUs on premises alongside cloud-based nodes on Microsoft Azure, with additional NVIDIA GPU capacity, including Blackwell systems, expected later as the effort scales. It is also using NVIDIA’s BioNeMo open developer platform and the MONAI framework in its research pipelines, according to the report.
C-BRAIN says some of the data it is trying to use has been difficult to access through conventional AI systems, including unpublished findings, negative results and governed proprietary data. The consortium refers to this as “dark data.” Dr. Ade Ojewole, C-BRAIN’s chief technology officer, said, “A lot of life sciences data doesn’t make its way into the frontier models,” adding that this leaves “a lot of intelligence on the table.”
The group says it has already built two open-source tools. OpenScientist, developed with Lawrence Berkeley National Laboratory in work led by research scientist Justin Reese, is described as an agentic AI co-scientist that can generate hypotheses, run analyses and produce auditable insights. InsightEngine, built with 387 Labs, is described as a platform that tests hypotheses against the scientific literature, analyzes user datasets, explores a biomedical knowledge graph and checks claims against evidence to reduce hallucinations before experiments are run.
In published case studies cited by the source, OpenScientist completed analyses in minutes that domain scientists estimated would take weeks to months, at a cost of under $10 per run. C-BRAIN says that, across more than 200 research groups, the time savings could translate into more than one million hours of research capacity annually at scale. That figure is presented by the consortium as a scenario-based estimate rather than a measured outcome.
Looking ahead, the consortium says it aims to build a multimodal biological foundation model using neuroimaging, multi-omics, fluid biomarkers and longitudinal clinical measures. Bateman said the objective is to accelerate discoveries that could support prevention and treatment development, while Ojewole said substantial optimization will be required as the group pushes beyond current spatial omics machine learning limits. The timing of any resulting scientific breakthroughs remains unknown.
