Surgical Reality Viewer

Surgical Reality Viewer is medical imaging visualization software that assists trained healthcare professionals with preoperative and intraoperative surgical planning. It displays 2D and 3D renderings of DICOM-compliant CT or MRI patient images, generates preliminary segmentations of normal anatomy using machine learning algorithms, and provides tools for manipulation and editing of segmentations and 3D models. The software is intended for adult patients aged 22 years and over, with the clinician responsible for all final patient management decisions.

Surgical Reality Viewer matches its predicate device (Ceevra Reveal 3) across all key technological characteristics: both use product code QIH under regulation 892.2050, support CT and MRI imaging modalities, are intended for healthcare professional users, provide interactive manipulation and 3D visualization, enable pre- and intra-operative viewing, include semi-automatic segmentation of supported normal anatomy, and do not provide real-time surgical guidance or quantitative outputs. The devices are functionally and technologically equivalent.

The device was developed in accordance with IEC 62304:2006/Amd 1:2015 (Medical device software lifecycle processes), IEC 81001-5-1 (Health software safety, effectiveness and security), and FDA guidance documents on premarket submissions for software and cybersecurity. Performance was verified using the Sørensen–Dice coefficient (DSC) to compare machine learning segmentation accuracy against ground truth segmentations.

Surgical Reality Viewer is substantially equivalent to Ceevra Reveal 3 based on identical indications for use, identical intended use in preoperative and intraoperative visualization of CT/MRI images, matching technological characteristics, and equivalent performance testing demonstrating high segmentation accuracy (DICE scores 0.84–0.97 across anatomical structures) and clinical usability scores (4.7–5.0). The device raises no new safety or effectiveness questions compared to the predicate.