IRISeg

IRISeg is a standalone software application that leverages machine learning and manual tools to process contrast-enhanced kidney CT images, automatically segmenting key kidney structures to generate virtual 3D models. This device is primarily intended for visual, non-diagnostic use in preoperative surgical planning and intraoperative display, aiming to enhance visualization for qualified clinicians.

IRISeg, a new Class II medical image management and processing system by Intuitive Surgical Inc., has been determined substantially equivalent to legally marketed devices, allowing its market entry under general regulatory controls. This standalone software processes contrast-enhanced kidney CT images to support preoperative surgical planning and intraoperative display, providing visual output that is non-diagnostic and requires review by qualified clinicians. The device integrates both manual and machine learning-enabled tools for image analysis and segmentation, specifically designed to automatically segment four key kidney structures: parenchyma, artery, vein, and collecting system, while requiring manual input for masses. Performance testing, including rigorous software verification, machine learning algorithm evaluation using 81 clinical kidney CT scans, and cybersecurity assessments, confirmed that IRISeg's design output meets requirements without raising new safety or effectiveness concerns.

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 manual segmentation tool compare in accuracy and efficiency to the machine learning algorithm for complex kidney pathologies not automatically segmented?; What specific clinical benefits and improvements in surgical outcomes are anticipated from using IRISeg for preoperative planning and intraoperative display?; Given its non-diagnostic use, what are the precise guidelines or recommendations for clinicians to integrate IRISeg's 3D models with diagnostic imaging for comprehensive patient assessment?; Could the machine learning algorithm be expanded in future iterations to include automated segmentation of renal masses or other pathological structures?; What were the specific metrics and thresholds used to evaluate the performance of the machine learning algorithm on the 81 clinical kidney CT scans, and how do these benchmark against current clinical standards?