Intended Use

The device is designed to aid the clinical assessment of adult chest x-ray cases with features suggestive of pneumothorax and tension pneumothorax in the medical care environment. It analyzes cases using an artificial intelligence algorithm to identify findings and outputs case-level results for worklist prioritization or triage in PACS or RIS. Intended for use by trained clinicians qualified to interpret chest x-rays as part of their scope of practice. Not intended as standalone diagnostic or to rule out pneumothorax.

Technology

Software integrates with RIS and PACS, uses a convolutional neural network trained on over 1.5 million chest x-rays from diverse datasets to detect pneumothorax and tension pneumothorax. Outputs case-level triage results to worklist software for prioritization. Uses deep learning with labeled data by radiologists. Allows configurable triage prioritization and does not downgrade existing priorities.

Performance

Performance evaluated in retrospective study of 949 chest x-rays with 413 positives (including 123 tension pneumo) and 536 negatives from multiple sites and imaging protocols. The device achieved high AUCs (0.979 pneumothorax, 0.988 tension pneumothorax), sensitivities and specificities above 80%, enabling clinically effective triage. Triage turnaround time averaged 20.57 seconds, comparable to predicate. Testing adhered to FDA performance guidelines for radiological triage software (21 CFR 892.2080).

Predicate Devices

No predicate devices specified

Device Timeline

  • 1

    Submission

    12/16/2021

    2 months
  • 2

    FDA Approval

    2/24/2022

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.