How a hospital network ran FDA-cleared chest X-ray AI in production across three facilities.
Get StartedA three-hospital deployment: PACS integration, clinical change management, and six-month outcomes.
Hospital radiology departments reading ~85,000 chest X-ray studies a year.
Three facilities carrying unfilled radiologist positions and rising workload.
AI output reaches the radiologist workflow on every eligible study, not a pilot subset.
Studies auto-route at acquisition and findings return as a DICOM Structured Report.
Dismissed findings are reviewed monthly and confidence thresholds tuned from real reads.
Scope your own clinical AI deployment with our healthcare engineering team.
A high-volume network needed faster, safer chest X-ray reads.
Over 90 days, more radiologists chose to see the AI overlay from the start of the read.
Urgent pneumothorax studies read first, regardless of acquisition time.
Findings missed on first read were confirmed or dismissed with justification.
Lower cognitive load on routine studies and in high-volume sessions.
Each radiologist chose when to see AI findings, so adoption grew.
Every number below was measured in production after launch — not projected in a pitch deck.
Urgent ED Report Time — Down from 47 minutes
Radiologists Report Positive Impact — At the six-month survey
Studies Processed Annually — Across all three facilities
Hospitals Live in Production — Network-wide rollout
