Demo

Genomics-aware CDS recommendations

Pick a synthetic sample to see how Engine 2's deterministic rule engine produces structured Recommendations from a parsed NGS report. Every output ships with a full evidence chain and a Non-Device CDS attestation.

Educational demonstration only. Not a medical device. Not for clinical decision-making. Output traces to publicly available evidence. Clinicians retain decision authority.

Try an example

How this view works

Recommendations are derived from the evidence-join endpoint: FDA-approved therapies become therapy-match cards, resistance-tagged CIViC evidence becomes resistance-warning cards, CPIC guidelines become PGx-dosing cards, ClinicalTrials.gov matches become trial-match cards. Engine 2 in production adds deterministic rule-engine logic for contraindications and combined-predicate triggers (e.g. BRCA + HRD → PARP) beyond what's rendered here.

How the CDS layer works under the hood

Deterministic rule routing across five categories — biomarker eligibility, drug-gene PGx, trial matching, contraindications, follow-up. Read full case walkthroughs: