BriefCase

The BriefCase for iPE Triage employs an artificial intelligence algorithm to analyze contrast-enhanced chest CT images, aiming to expedite the detection and notification of suspected incidental pulmonary embolism for improved radiology workflow prioritization. It functions as an adjunctive tool operating in parallel with standard care, providing informational alerts to radiologists without making diagnostic claims or altering primary image data.

BriefCase for iPE Triage is a radiological computer-assisted triage and notification software designed to assist hospital networks and trained radiologists in workflow management. It utilizes an artificial intelligence algorithm to analyze contrast-enhanced chest CT images (excluding dedicated CTPA protocols and restricted to single-energy exams from GE and Siemens scanners) to identify suspected cases of Incidental Pulmonary Embolism (iPE). The system architecture includes an Aidoc Hospital Server for image acquisition, an Aidoc Cloud Server for image processing, and an Aidoc Worklist Application installed on the radiologist's desktop. This application provides notifications for flagged studies, accompanied by compressed, non-diagnostic preview images, which are intended solely to prompt earlier review of the original images within the Picture Archiving and Communication System (PACS), thereby enabling faster triage and prioritization in parallel with the standard imaging interpretation workflow. The device's efficacy was assessed in a retrospective, blinded, multicenter study using 268 cases, of which 74 were positive for iPE. The software achieved a sensitivity of 90.5% and a specificity of 88.7% for iPE detection. A significant finding was the potential for expedited case review; the BriefCase time-to-notification for true positive iPE cases averaged 4.7 minutes, markedly faster than the standard of care time-to-exam-open of 223.3 minutes. This suggests a potential mean time saving of approximately 220.9 minutes, allowing radiologists to address critical findings more rapidly. The study also reported a Negative Predictive Value (NPV) of 99.7% and a Positive Predictive Value (PPV) of 17.6%. The device does not alter original medical images and is not intended for diagnostic use.

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 AI algorithm's performance for iPE detection on general contrast-enhanced chest CTs compare to its performance on dedicated CTPA scans, if such a comparison were feasible with this algorithm?; What is the real-world impact on radiologist workflow and potential for notification fatigue given a Positive Predictive Value of 17.6% for iPE?; What are the specific image features or patterns the AI algorithm has learned to identify as indicative of incidental pulmonary embolism in non-CTPA chest CTs?; How does the device's performance (sensitivity, specificity, time-saving) vary when used with CT scanners from manufacturers other than GE and Siemens, or with dual-energy CT exams?; What is the minimum contrast enhancement level required for the AI algorithm to reliably detect iPE, and how does this compare to typical non-dedicated chest CT protocols?