Rapid LVO

Rapid LVO is a radiological computer-aided triage and notification software that utilizes a software algorithm to analyze CTA head images for suspected Large Vessel Occlusions in the ICA or MCA-M1, aiming to expedite workflow prioritization for potential stroke patients. The system demonstrated high sensitivity and specificity in performance testing, providing rapid notifications to clinicians while operating in parallel to standard diagnostic procedures.

Rapid LVO is a radiological computer-aided triage and notification software designed for the analysis of CTA head images. Its primary purpose is to assist hospital networks and trained radiologists in optimizing workflow triage by automatically flagging and communicating suspected positive Large Vessel Occlusion (LVO) findings in the internal carotid artery (ICA) or the M1 segment of the middle cerebral artery (MCA-M1). The system operates in parallel with standard image interpretation, analyzing images using a software algorithm and presenting notifications of suspected LVOs to users via a server or standalone desktop application. These notifications include compressed preview images intended for informational purposes only to facilitate quicker prioritization, while the original medical images remain unaltered and the device is not intended for diagnostic decisions. The device employs traditional programming algorithms integrated within the Rapid Platform, which handles DICOM processing, job management, and imaging module execution. It is not a diagnostic tool; its results are intended to be used with other patient information and professional judgment for triage, with clinicians responsible for viewing full, non-compressed images for diagnosis. Performance validation on 217 scans demonstrated high accuracy, with a sensitivity of 0.96 (95% CI: 0.91-0.97) and specificity of 0.98 (95% CI: 0.93-0.99) for LVO detection, a Positive Predictive Value of 0.98, a Negative Predictive Value of 0.96, and an RoC AUC of 0.99. The average notification time was 3.18 minutes, and performance was validated across different demographics, scanner manufacturers, and image slice thicknesses.

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 Rapid LVO's software algorithm differentiate between true Large Vessel Occlusions and potential mimics, such as severe stenosis or artifacts, within CTA head images?; What is the variability in Rapid LVO's performance when processing CTA scans acquired with different contrast media protocols or from patients with atypical cerebrovascular anatomy?; What were the primary characteristics of the false positive and false negative cases identified during the performance validation studies, and what measures are in place to mitigate these?; How does the 3.18-minute average notification time for suspected LVOs by Rapid LVO translate into measurable improvements in door-to-intervention times in real-world clinical settings?; Given that the MCA-M1 definition includes post-bifurcation M1 segments, how does the algorithm's sensitivity and specificity compare for occlusions in these distal M1 segments versus more proximal M1 occlusions?