BriefCase

BriefCase is an AI-powered software designed to analyze chest and abdominal CT images for the detection of suspected Vertebral Compression Fractures, aiming to improve radiology workflow triage. It operates by processing DICOM images through a cloud-based algorithm and notifying clinicians of potential positive cases via a desktop application, demonstrating high sensitivity and AUC in performance studies.

BriefCase, developed by Aidoc Medical, Ltd., is a radiological computer-aided triage and notification software designed to analyze chest and abdominal CT images. Its purpose is to assist hospital networks and trained medical specialists in the bone health setting by improving workflow triage through the flagging and communication of suspected positive cases of Vertebral Compression Fractures (VCFx). The system employs an artificial intelligence algorithm and consists of three modules: an Aidoc Hospital Server for image acquisition, an Aidoc Cloud Server for image processing, and an Aidoc Desktop Application for workflow integration, which displays a feed of suspect cases. DICOM images are acquired, filtered, de-identified, and processed chronologically. Upon detecting a potential VCFx, the system notifies users with a compressed, low-quality, unannotated preview image intended solely for prioritization and explicitly not for diagnostic use, operating in parallel with standard image interpretation workflows without altering original images or the standard reading queue. The performance of BriefCase was assessed in a retrospective, blinded, multicenter study involving 318 cases from five U.S. clinical sites, with ground truth determined by three senior board-certified radiologists. The primary endpoint, the area under the receiver operating characteristic curve (AUC), was 97.6%, surpassing the pre-specified goal of >0.95. The software achieved a sensitivity of 95.11% and a specificity of 93.28% for VCFx detection. Secondary metrics included a mean time-to-notification of 117.2 seconds for true positive cases. Additional reported values were a Positive Predictive Value (PPV) of 55.2% and a Negative Predictive Value (NPV) of 99.5%. The study cohort had a mean age of 65.7 years, comprised 46.5% males and 53.5% females, and involved CT scans from various major manufacturers.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the algorithm's performance (sensitivity, specificity, AUC) vary across different CT scanner manufacturers and imaging protocols?; What specific features or image characteristics does the AI algorithm primarily rely on to identify suspected Vertebral Compression Fractures?; Given the PPV of 55.2%, what is the anticipated impact on radiologist workload in terms of reviewing flagged cases that are ultimately negative, and are there strategies to optimize this?; How does the time-to-notification of BriefCase (117.2 seconds) compare to the actual time saved in clinical practice for initiating VCFx patient management, considering the entire workflow?; What were the inclusion and exclusion criteria for the 318 cases in the pivotal study, and how might selection bias influence the reported performance metrics?