BunkerHill BMD is an AI-based software tool designed to estimate bone mineral density from existing CT scans of the spine in adults aged 30 and over. It helps clinicians identify low bone density without requiring dedicated bone densitometry scans, supporting retrospective assessment and aiding clinical decision-making.
The Bunkerhill BMD Algorithm is a post-processing AI-powered software intended for adults 30 years and above to assess estimated DXA-measured average areal bone mineral density of spinal bones from existing CT scans and outputs a flag for low bone density below a pre-specified threshold. It is not intended to replace DXA or any other tests dedicated to BMD measurement. It enables retrospective assessment of bone density from CT scans acquired for other purposes or in conjunction with other medically appropriate CT procedures, and assessment without a phantom as an independent measurement procedure.
The device is a software-only medical device using deep-learning computer vision and post-processing algorithms to estimate bone mineral density from CT images in DICOM format. It serves as an assistive tool providing quantitative BMD assessments and generates a report. It shares the same technological characteristics as its predicate device, including automatic measurement and averaging of Hounsfield Units in spinal bones.
Performance was validated retrospectively on 371 CT studies from four diverse sites, showing sensitivity of 81.0% and specificity of 78.4%, passing acceptance criteria with 95% confidence intervals above 70%. Additional metrics included Pearson correlation of 0.791, AUROC of 0.883, PPV 73.6%, and NPV 84.8%. Device performance was consistent across patient subgroups, CT manufacturers, slice thickness, and reconstruction kernel, demonstrating generalizability.
No predicate devices specified
Submission
8/2/2024
FDA Approval
4/8/2025
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.