Viz ANEURYSM, Viz ANX

Viz ANEURYSM is an AI-powered software designed to assist radiologists by rapidly analyzing CTA head images for suspected aneurysms (≥4mm) and prioritizing these cases for review via a mobile application. Its methodology relies on a machine learning algorithm to identify aneurysmal features, providing a non-diagnostic triage tool that significantly reduces notification times and demonstrates high sensitivity and specificity in clinical studies.

Viz ANEURYSM (Viz ANX) is a radiological computer-assisted triage and notification software specifically engineered for the analysis of Computed Tomography Angiography (CTA) images of the head. Its core function is to enhance workflow efficiency for hospital networks and trained radiologists by automatically flagging and prioritizing studies that exhibit suspected intracranial aneurysms measuring 4mm or greater in diameter. The system utilizes an advanced artificial intelligence (AI) and machine learning (AI/ML) algorithm. This algorithm processes CTA scans, which are automatically transmitted to Viz.ai’s backend server for analysis. Upon identifying a potential aneurysm, Viz ANEURYSM alerts users by highlighting the case within a dedicated standalone mobile application, Viz VIEW. This mobile platform presents compressed, non-diagnostic preview images, enabling clinicians to rapidly assess and prioritize cases for further detailed review using standard diagnostic tools, all while operating in parallel with existing standard of care image interpretation workflows and without altering the original medical data. The scientific basis of Viz ANEURYSM lies in its sophisticated AI algorithm's capacity to discern subtle image features indicative of aneurysms from CTA data. Clinical performance evaluations, conducted on a dataset of 315 scans, demonstrated robust efficacy: the software achieved a sensitivity of 93% and a specificity of 89% for detecting aneurysms of 4mm or larger, successfully meeting pre-defined performance benchmarks. The Area Under the ROC Curve (AUC) was calculated at 0.967, underscoring the classifier's strong discriminatory power. A key operational benefit highlighted was the significant reduction in time-to-notification; Viz ANEURYSM provided alerts in an average of 3.67 minutes, markedly faster than the 43.6-minute average observed in standard care pathways for the cases analyzed. It is designed as an adjunctive tool to support, not replace, radiologist expertise; diagnostic decisions remain the purview of radiologists using original images, and the software is limited to imaging data analysis for triage purposes.

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 differentiate between various types of intracranial aneurysms (e.g., saccular, fusiform) or features that might mimic aneurysms, especially for those near the 4mm detection threshold?; What was the variability in performance (sensitivity/specificity) observed across different CTA scanner models and acquisition protocols beyond the summarized manufacturer data, and are there specific scan parameters that optimize the algorithm's detection capabilities?; Given the significant time-to-notification improvement, what is the quantifiable impact on patient outcomes or downstream clinical decision-making in real-world hospital network settings where Viz ANEURYSM is implemented?; How does the algorithm's performance adapt to evolving aneurysm characteristics, such as growth, rupture, or post-treatment changes, if follow-up scans are processed through the system?; What are the computational load and data throughput requirements for the backend server processing, and how does the system scale to accommodate the imaging volume from a large, multi-site hospital network?