BriefCase-Triage

BriefCase-Triage is a radiological software that uses artificial intelligence to analyze CT chest, abdomen, or chest/abdomen exams with contrast in adults aged 18 and older. It flags suspected cases of Aortic Dissection and sends notifications to assist radiologists in workflow prioritization, while operating in parallel to standard image interpretation without altering original images.

Both the subject and predicate devices are deep learning AI algorithms that process DICOM-compliant CT images and provide notifications with compressed, low-quality grayscale preview images for triage purposes. The subject device differs primarily in algorithm training process and performance metrics, with four additional operating points compared to the predicate's two, and demonstrates improved time-to-notification (10.7 seconds vs. 38.0 seconds).

Not stated in this summary.

Both devices raise identical safety and effectiveness questions regarding accurate triage of findings. They share the same fundamental design—deep learning algorithms for image analysis, notification-only alerts, compressed preview images marked not for diagnostic use, and parallel workflow operation that does not remove or de-prioritize cases from standard reading queues. The subject device achieved sensitivity of 92.7% and specificity of 92.8% with 95% confidence intervals exceeding the 80% performance goal, matching the predicate's clinical approach to preemptive triage while delivering comparable or better time-to-notification benefits.