
What happened
nference and Cedars-Sinai announced a long-term agreement on November 20 to build a federated AI research platform using over a decade of de-identified records from more than 5 million patients.
Why it matters
The platform is intended to span clinical notes, labs, radiology and pathology imaging, genomics, waveforms and vitals, with stated uses in drug discovery, clinical trial design and future diagnostics — though the parties disclosed neither contract duration nor financial terms.
WHO IT HITSHospital research and informatics teams and life-sciences researchers are the likely users of a shared multimodal dataset of this kind, if the intended capabilities are delivered. Neither party disclosed contract terms, so the near-term operational impact on either organization's staff remains unclear.
Summaries like this, every morning.
The agreement pairs two different kinds of assets. nference brings the federated software architecture, multimodal data harmonization, and physician-adjudicated curation; Cedars-Sinai brings the de-identified clinical records themselves — over a decade's worth, from more than 5 million patients. The stated breadth of data types runs from clinical notes and laboratory results through radiology and pathology imaging, genomics, patient-consented biospecimens, electrophysiology waveforms, vitals, and flowsheets, refreshed continuously as routine care progresses.
Both sides framed the effort around connecting information across organs, clinical modalities, and time. Venky Soundararajan, nference's founder and co-CEO, said understanding health and disease requires connecting what is happening across organs, across clinical modalities, and across decades of each patient's medical journey. Sumeet Chugh, Cedars-Sinai's vice dean and chief artificial intelligence health research officer, said the collaboration is expected to give clinicians and scientists "new ways to learn from the collective experience of patient care." The stated research uses span drug discovery, clinical trial design, and future diagnostics.
What the announcement does not include is as notable as what it does: neither party disclosed contract duration or financial terms, and the platform remains under development, with capabilities still described as "intended" rather than delivered, according to unite.ai. That places the agreement at the stage of a declared build rather than a live capability, with the commercial and operational details still to emerge.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.