Boston Children's Hospital has used OpenAI technology to help diagnose over 40 complex rare disease cases, according to a case study published by OpenAI. The hospital built a system using OpenAI's models to act as an analytical co-pilot for medical teams working on diagnostic odysseys that often span years.
How the AI Assistant Works
Per OpenAI's case study, the system processes unstructured patient notes, lab results, and genomic data, cross-referencing them against medical literature to surface potential diagnoses and generate summaries for physician review. The intent is not to replace clinicians but to reduce the time spent manually sifting through scattered data sources.
Boston Children's also reports using the technology for administrative tasks, including drafting insurance pre-authorizations and summarizing patient histories for handoffs between care teams, aiming to reduce the paperwork burden on staff.
Why It Matters
The case study offers a concrete example of large language models being applied in a clinical setting, beyond the research stage. It suggests a template other healthcare institutions might follow to combine diagnostic support with reductions in administrative workload, though OpenAI's own case study is the sole source of these figures and outcomes have not yet been independently verified by outside researchers.