Sure for Sure
Sure for Sure turns clinical conversations and the patient's record into a focused work queue: ask the patient, reconcile the chart, or suppress an already-settled issue.
Core trick
Turns each patient medication statement into a claim, compares asserted fields against same-patient FHIR evidence, and routes unresolved conflicts to review.
Tags
Description
Sure for Sure turns clinical conversations and the patient's record into a focused work queue: ask the patient, reconcile the chart, or suppress an already-settled issue. It weighs how confidently the patient asserts each claim against what the record supports, catching treatment changes and record conflicts that a clean note can flatten — surfacing what needs a clinician's attention without adding alert fatigue.
README excerpt
# Sure for Sure Medication reconciliation fails in the gap between what a patient says and what the record can actually establish. Sure for Sure turns each patient medication statement into an inspectable claim, links it to same-patient FHIR evidence, compares every asserted field, detects later clarification, and routes the uncertainty that remains. Originally conceived during a healthcare AI hackathon, Sure for Sure was rebuilt from the ground up as a tested, provenance-aware medication reconciliation and clarification pipeline.  ## End-to-end workflow ```text transcript → exact patient claim + confidence → same-patient FHIR/longitudinal medication evidence → deterministic identity linking → field-level support / silence / contradiction / source conflict → CED attention score → disputed-field resolution from later patient speech → patient clarification, chart cleanup, clinician review, or no action ``` The authoritative runtime is the top-level [`pipeline/`](pipeline/) package, used by both the [`backend/`](backend/) CLI/API and the [`frontend/`](frontend/) review app. The ten explicit calls are visible in [`pipeline/runner.py`](pipeline/runner.py). CED is deliberately simple: ```text CED = patient-confidence signal × record-divergence signal ``` It is an attention heuristic, not a calibrated probability, clinical-risk score, or assertion that the chart is more truthful than the patient.