Lunit INSIGHT CXR Triage
Lunit Inc.Lunit INSIGHT CXR Triage is an AI-based software tool designed to analyze adult chest X-ray images and identify potentially critical lung conditions such as pleural effusion and pneumothorax. It flags suspicious cases in the radiology workflow, enabling radiologists to prioritize urgent cases more efficiently without altering the standard reading queue. It operates as a supportive notification system and is not intended for standalone clinical decision making.
Intended Use
Lunit INSIGHT CXR Triage is a radiological computer-assisted triage and notification software that analyzes adult chest X-ray images for the presence of pre-specified suspected critical findings (pleural effusion and/or pneumothorax). It uses artificial intelligence to analyze images and provides case-level output available in the PACS/workstation for worklist prioritization or triage. The software does not send proactive alerts directly to medical specialists and is not intended for standalone clinical decision-making.
Technology
Lunit INSIGHT CXR Triage is a software-only AI-based radiological computer-assisted prioritization tool. It receives chest X-ray DICOM images from PACS or radiology imaging equipment, de-identifies and analyzes them for pleural effusion and pneumothorax using AI algorithms. The software flags positive cases in the PACS or workstation worklist for prioritization. It provides passive notification and tech-level notification post user interpretation. It operates in parallel with and does not alter the standard clinical workflow. It supports deployment on several computing platforms including PACS, radiology equipment, or cloud.
Performance
The device was validated with clinical and nonclinical tests showing high performance detecting pleural effusion and pneumothorax on chest X-rays. Nonclinical testing on 1,385 images showed ROC AUC around 0.99 for both conditions, with sensitivity and specificity above 94%. Clinical pivotal studies on 1,708 chest radiographs (NIH and multisite India datasets) yielded ROC AUCs approximately 0.96-0.97 with sensitivity and specificity above 88% for both conditions. Device processing time averaged around 20 seconds per case, comparable to cleared predicate devices. Data supports the device's safety and effectiveness for triage and prioritization use.
Predicate Devices
No predicate devices specified
Device Timeline
- 1
Submission
6/4/2021
5 months - 2
FDA Approval
11/10/2021
Other devices from Lunit Inc.
- Lunit INSIGHT DBT v1.12024 • 510K
- Lunit INSIGHT DBT2023 • 510K
- Lunit INSIGHT MMG2021 • 510K
Ready to Sharpen Your Edge?
Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.
We respect your privacy. Unsubscribe at any time.