Onqi Screening: Detect Early. Refer Fast.
An AI-powered screening engine designed to help clinics identify at-risk patients early — and generate revenue through value-based care programs.
Proven Outcomes for Clinics
2–3× More Patients Identified
NLP flags at-risk patients CMS often misses — using unstructured notes, not just checkboxes.
30% Faster Referrals
Prefilled docs and one-click referrals reduce time-to-referral delays by 30%+.
Boosted Reimbursement ROI
Clinics unlock value-based bonuses through ACO REACH and UDS — while closing care gaps.
Designed for Every Team Member
Clinicians
View risk flags directly in the chart — no digging, no surprises.
Care Managers
Auto-generate compliant referrals in seconds, fully documented.
VBC Executives
Track performance, trends, and ROI across your screening program.
Architecture Walkthrough
Each module maps to a clinical workflow challenge — and is built for modular deployment across diverse EHR environments.
Patient Data Intake
Accepts structured (FHIR API) and unstructured (CSV uploads) data — enabling rapid integration with diverse EHRs.
NLP Risk Extraction
Uses ScispaCy + MedSpaCy to identify lung cancer risk factors (pack-years, secondhand exposure, job risks) from clinical notes.
Eligibility Rules Engine
Applies USPSTF/NCCN guidelines through configurable logic to flag patients — even those not CMS-eligible — who may benefit from screening.
Backend & Data Layer
FastAPI services + PostgreSQL database deployed via Cloud Run and Cloud SQL for modular, autoscaling infrastructure.
Referral Generator + CMS Docs
Jinja2 templates + WeasyPrint to produce compliant referrals and documentation — instantly available to clinicians via a simple UI.
System Overview Diagram
Referral Output Example
CMS-ready referrals generated in seconds — complete with risk factors, shared decision-making text, and compliance metadata.
View Sample Referral (PDF)