Leiden University Medical Center

Radiologists face the risk of missed findings in CT chest angiography studies, particularly for pulmonary embolism — a potentially life-threatening condition. As a teaching medical institution, LUMC sought advanced tools to support radiologists in catching subtle, hard-to-detect cases while maintaining an unobtrusive workflow.

LUMC deployed Oxipit's CT PE Quality application, an AI-powered second-reader tool integrated via Sectra's radiology imaging platform. After a radiologist submits a report, the AI checks it against its own findings; if potential missed pulmonary embolism findings are identified, the study is flagged for secondary review and an automated notification is sent to the reporting radiologist. The deployment complements LUMC's existing Oxipit suite, which includes Quality tools for chest and MSK X-rays.

The CT PE Quality application is currently used in a research capacity at LUMC. Radiologists have rated the Quality tools highly for operating in a seamless, unobtrusive manner. The broader Oxipit suite, including ChestLink, enables LUMC to automate healthy patient chest X-ray reporting and reduce radiologist workload and reporting backlogs.

• AI as a second reader — rather than a first reader — can integrate smoothly into existing radiologist workflows without disruption.
• Combining autonomous reporting (ChestLink for normal studies) with quality assurance tools (CT PE Quality) creates a layered AI strategy that addresses both throughput and diagnostic accuracy.
• Deploying via an established imaging platform (Sectra) accelerates integration and adoption in complex hospital environments.