Color Health built an LLM-powered 'Cancer Copilot' to assist with the workup of patients newly diagnosed with cancer. Color reports more than 95% concordance with guidelines on a dataset of anonymized real-world patients at UCSF Health.
Reported outcomes
Outcomes are reported exactly as stated by the source; we have not independently audited them.
How it works
- Extracts explicit decision factors from the patient record.
- Applies declarative guideline logic to recommend workup steps.
- Produces an explanatory summary with a reasoning chain and guideline references, designed for expert review.
Risks & compliance
- Clinical decision support; not medical advice
- Evaluation was on a retrospective dataset
Healthcare cases cover operational and commercial workflows. Nothing here is medical advice or a claim about a device's or drug's efficacy.
Tools mentioned & how to build it
Disclosure: Some links may earn us a commission; it never affects grades. Tools are listed by fit to this case, never by payment.
Build a version yourself
- Lindy
No-code AI agents for inbox, scheduling and sales admin. - Fireflies.ai
AI meeting notes that sync summaries and action items to your CRM.
Sources
“The system achieved a high degree of concordance with guidelines (> 95%) on a data set composed of anonymized real-world patients at UCSF Health.”Expertise Is What We Want
Grade rationale: Named organization reporting a quantified result on its own blog. Automated check on Oct 17, 2026: 1/1 quotes and 1/1 outcome claims matched the source. Compiled by the Agent Casefile research desk (AI-assisted; see methodology). Spot an error? Request a correction.
