FDA-Cleared AI Device

Lunit INSIGHT CXR

Lunit

Detects and localizes several abnormality classes on chest radiographs, including nodules, consolidation, pneumothorax, and pleural effusion.

At A Glance

Manufacturer
Lunit
Medical specialty
Radiology
Clearance route
510(k)
Year authorized
2021 onward
Modality
Chest X-ray
Role in workflow
Concurrent read

Multi-finding chest X-ray models are harder to evaluate than single-finding ones, because aggregate accuracy hides wide variation between classes. A model can look strong overall while performing poorly on the one finding a particular department cares about.

Where It Is Used

  • Chest radiograph interpretation support
  • Health screening programs
  • Emergency department chest film review

Limitations Worth Knowing

Ask for per-finding performance, not aggregate AUC. Nodule detection and subtle consolidation are typically the weakest classes, and they are frequently the reason a department wanted the tool.

Regulatory status, indications, and labeling change over time. Confirm current status on the FDA's AI/ML-Enabled Medical Devices list and in the manufacturer's current instructions for use before relying on anything stated here.

← Back to FDA-Cleared AI Devices in Radiology