COIOS
Emergent technology Item new preprint

A generative model of patient records stratifies five-year risk of first cancer

A preprint describes an autoregressive generative model trained on electronic health records from millions of patients, reporting that supervised adaptation improved prediction of a first cancer diagnosis within five years across five cohorts.

The authors trained an autoregressive generative model on longitudinal electronic health records from millions of patients, explicitly representing the irregular time gaps between encounters, then adapted it with parameter-efficient supervised fine-tuning for pan-cancer risk stratification. Across five large EHR cohorts, the adapted model improved prediction of a first cancer diagnosis within a five-year window relative to the foundational representation alone; no absolute discrimination figures are given in the abstract. The authors frame the work as retrospective evidence supporting prospective evaluation for prioritising patients for risk-based screening, including pancreatic and ovarian cancer.

Why it is interesting: Tests whether a general-purpose generative model of patient histories, rather than a task-specific risk equation, can identify who to screen, so far only retrospectively.

Source
medRxiv, 19 September 2026
DOI
10.64898/2026.09.09.26362676
Type
Preprint
Design
Retrospective development and validation of a generative EHR foundation model (GenEHR) with supervised adaptation, five large EHR cohorts, millions of patients, five-year horizon
Verdict
New finding
Driver
AI in medicine