EFAI CARDIOSUITE CTA ACUTE AORTIC SYNDROME ASSESSMENT SYSTEM

The EFAI CARDIOSUITE CTA ACUTE AORTIC SYNDROME ASSESSMENT SYSTEM is an AI-powered software tool that analyzes adult chest or chest-abdomen CTA scans to identify suspected aortic dissection or intramural hematoma. Its purpose is to facilitate workflow triage and prioritization for radiologists by providing case-level notifications of potential acute aortic syndrome cases to PACS/workstation systems.

The EFAI CARDIOSUITE CTA ACUTE AORTIC SYNDROME ASSESSMENT SYSTEM (EFAI AASCTA) is a radiological computer-aided triage and notification software designed for analyzing chest or chest-abdomen Computed Tomography Angiography (CTA) scans in adults aged 22 and older. Its primary purpose is to assist hospital networks and medical specialists by improving workflow triage. The system employs an artificial intelligence algorithm, specifically deep learning techniques, to automatically identify images with features suggestive of acute aortic syndrome, namely aortic dissection (AD) or aortic intramural hematoma (IMH). Upon detection of suspected positive cases, it flags these studies and communicates this information to a PACS/workstation, thereby enabling earlier review of critical cases by radiologists. The EFAI AASCTA system is intended as an adjunctive tool to aid in the prioritization of radiological medical images; it does not mark or highlight specific anomalies on the images, nor are its results meant for standalone diagnostic decision-making or to rule out acute aortic syndrome. A retrospective, blinded, multisite clinical validation study involving 380 CTA studies demonstrated the system's performance. The EFAI AASCTA achieved a sensitivity of 0.929 and a specificity of 0.915 in identifying AD or IMH, with an average system processing time of approximately 37.86 seconds per study. Subgroup analyses indicated consistent performance across different genders, age groups, CT manufacturers, slice thicknesses, and in the presence of image quality issues or other confounding radiological findings.

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. What specific deep learning architecture is employed by the EFAI AASCTA, and how was its training dataset curated to ensure robustness against diverse patient demographics and image acquisition variability?; How does the system's performance (sensitivity/specificity) vary specifically between aortic dissection and aortic intramural hematoma, and are there plans to differentiate subtypes of dissection (e.g., Stanford Type A vs. Type B)?; What were the primary causes of false positives and false negatives in the clinical validation study, and what strategies are being considered to mitigate these?; Beyond notification, what is the average time saved in the diagnostic workflow from image acquisition to radiologist review for suspected positive cases when using EFAI AASCTA in a real-world clinical setting?; Are there any known limitations or decreased performance in patients with pre-existing aortic conditions, such as aneurysms or significant atherosclerosis, that might mimic or obscure signs of acute aortic syndrome?