BriefCase-Triage

BriefCase-Triage is radiological computer-aided triage and notification software that uses artificial intelligence to analyze contrast-enhanced CT images of the lungs in adults age 18 and older. It identifies suspected cases of incidental pulmonary embolism (iPE) and alerts clinicians with compressed preview images for workflow prioritization, operating in parallel to standard clinical care without altering original images.

The subject device consists of an AI algorithm module integrated with an image communication platform (separated from prior monolithic architecture), processes DICOM images, supports both single and dual-energy exams with slice thickness 0.5-5.0mm, and includes four additional operating points beyond the default. The predicate only supported single-energy exams with 0.5-3.0mm slice thickness and had no additional operating points.

Not stated in this summary.

Both devices are deep-learning AI algorithms that operate in parallel to standard workflow without removing or de-prioritizing cases from the reading queue. Both provide unannotated, compressed, low-quality preview images with explicit labeling against diagnostic use. The pivotal study demonstrated sensitivity of 91.7% and specificity of 91.4% (exceeding the 80% goal), with time-to-notification of 40.2 seconds comparable to the predicate's 282 seconds. The expanded inclusion criteria (dual-energy exams, thicker slices) and additional operating points do not raise new safety or effectiveness questions. Both devices achieve identical intended clinical functions and raise the same triage accuracy concerns.