IRISeg

IRISeg is a software application that processes DICOM-compliant MR or contrast-enhanced CT images to provide manual and machine learning-enabled tools for image segmentation. It generates output files that can be rendered as 3D models for preoperative surgical planning and intraoperative display. The output is intended for visual, non-diagnostic use and must be reviewed by clinicians responsible for patient management decisions.

The subject device adds support for DICOM-compliant MR images with manual segmentation capability for kidney, prostate, and rectal structures, whereas the predicate supported only CT images. The machine learning auto-segmentation algorithm for kidney CT structures remains unchanged. Both devices run on general-purpose computer hardware and use the same graphical user interface design.

IEC 62304 Edition 1.1 2015-06 Consolidated Version for software development; ISO 14971 for risk management; AAMI CR34971 for guidance on artificial intelligence and machine learning; FDA Guidance on Cybersecurity in Medical Devices (June 27, 2025) and Transparency for Machine Learning-Enabled Medical Devices (June 2024).

The subject device is substantially equivalent to predicate K242461 because it maintains the same intended use, indications for use, general design, and operational principles for CT segmentation. The addition of MR image support with manual-only segmentation does not introduce new risks or questions about safety and effectiveness, as it uses equivalent manual segmentation tools and the unchanged ML algorithm demonstrates equivalent performance.