EFAI Chestsuite XR Malpositioned ETT Assessment System (ETT-XR-100)

The EFAI Chestsuite XR Malpositioned ETT Assessment System is an AI-powered radiological computer-aided triage software designed to assist medical specialists by identifying suspected cases of vertically malpositioned endotracheal tubes relative to the carina on adult chest X-ray images. Utilizing deep learning algorithms for workflow prioritization, this device has demonstrated satisfactory sensitivity and specificity in clinical validation, supporting its safety and effectiveness for FDA marketing clearance.

Ever Fortune.AI, Co., Ltd. has received a determination from the U.S. Food & Drug Administration regarding their EFAI Chestsuite XR Malpositioned ETT Assessment System (ETT-XR-100). This system, categorized as radiological computer-aided triage and notification software, has been deemed substantially equivalent to legally marketed devices. This finding allows the company to market the device, provided they adhere to general regulatory controls, including annual registration, device listing, good manufacturing practices, and truthful labeling. The EFAI Chestsuite XR Malpositioned ETT Assessment System is intended to assist medical specialists and hospital networks by analyzing adult chest X-ray images. It uses deep learning algorithms to flag suspected cases of vertically malpositioned endotracheal tubes relative to the carina, facilitating workflow prioritization. While providing case-level outputs, it is not meant to direct attention to specific image areas, nor is it for standalone clinical decision-making. Clinical validation studies have confirmed its satisfactory performance in terms of sensitivity and specificity, demonstrating its safety and effectiveness across diverse patient populations and various clinical circumstances.

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 architectural details of the deep learning model contribute to its ability to accurately identify ETT malpositioning relative to the carina?; Could the detailed results (e.g., AUC, specific sensitivity/specificity values with confidence intervals) from the clinical validation studies be provided to evaluate the system's robustness?; How does the system differentiate between subtle malpositions and tubes that are correctly placed but at the edge of the acceptable range?; What is the expected impact on clinician workload and time-to-diagnosis or intervention in a real-world clinical setting?; Does the system's performance vary based on image quality, patient demographics, or the presence of other pathologies in the chest X-ray?