Bunkerhill Abdominal Aortic Quantification (AAQ)

Bunkerhill AAQ is a software-only medical device that uses deep learning algorithms to automatically measure the maximum abdominal aortic diameter from CT scans of the abdomen/pelvis in adults aged 22 and older. The device analyzes CT images with or without IV contrast that include the L1–L5 region of the abdominal aorta and assists radiologists by providing measurement output. Results are presented as an annotated preview image but are explicitly not intended for standalone clinical decision-making; physicians must review full images and clinical information.

Both the subject device (Bunkerhill AAQ) and predicate device (Briefcase Quantification) use deep learning algorithms with image databases, analyze non-gated CT images in DICOM format, serve as adjunctive support tools, and produce annotated preview images with original unmarked series retained in PACS. The minor technological difference is that Bunkerhill AAQ accepts both contrast and non-contrast CT scans, whereas the predicate accepts only contrast scans. Both are cloud-hosted, output DICOM-formatted results to PACS, and do not interfere with standard workflow.

Not stated in this summary. The document references 'Software Development and Validation & Verification Process' and FDA guidance on 'Content of Premarket Submissions for Device Software Functions,' but does not cite specific ISO, IEC, ASTM, or other consensus standards by number.

The devices are substantially equivalent because they share identical regulatory classification, product code, and core technological approach (AI-based diameter measurement from CT). Both serve the same adjunctive clinical purpose, target the same patient population and anatomical location, use comparable algorithms, and produce functionally equivalent outputs. The expansion to non-contrast scans is a minor input variation that does not alter the fundamental diagnostic use or principles of operation. Robust performance validation demonstrated mean absolute error of 1.58 mm across diverse patient subgroups (age, sex, aneurysm size, manufacturers, contrast types, and sites), establishing that the algorithm performs comparably to the predicate across clinically relevant scenarios.