uMR Ultra

The uMR Ultra is a 3.0 Tesla superconducting magnetic resonance imaging (MRI) system designed to produce sagittal, transverse, coronal, and oblique cross-sectional images as well as spectroscopic images for diagnostic visualization of anatomical structure and function of the head, body, and extremities. Trained physicians interpret these images and derived physical parameters to assist in diagnosis, with optional contrast agent use depending on the region of interest.

The uMR Ultra features a 70 cm patient bore (versus 75 cm on the predicate uMR Omega), maximum gradient amplitude of 100 mT/m (versus 45 mT/m), up to 192 receive channels (versus 96), and includes several new RF coils and accessories such as UHD SuperFlex coils, Tilt Support, Positioning Couch-top, Coil Support, and Patient Bore Projector. New software features include uVision Body Part Recognition, MQD, EasyBolus, Auto Bolus Tracker, Inline CEST, t-ACS, AiCo, EasyRegister, and enhanced inline cardiac function with ED/ES phase recognition.

Electrical Safety and EMC: ANSI/AAMIES60601-1, IEC 60601-1-2, IEC 60601-2-33, IEC 60825-1, IEC 60601-1-6, IEC 62304, IEC 62464-1. Image Quality: NEMA MS 1–9, NEMA MS 14, NEMA MS 19. Software: NEMA PS 3.1–3.20 (DICOM), FDA guidance on premarket submissions for software, cybersecurity. Biocompatibility: ISO 10993–5, ISO 10993–10, ISO 10993–23. Risk Management: ISO 14971. Quality System: 21 CFR Part 820.

The uMR Ultra is substantially equivalent to the uMR Omega predicate because it maintains the same fundamental magnetic field strength (3.0T), magnet type (superconducting), resonant frequency, number of transmit channels, and indications for use. Although the patient bore is smaller (70 vs 75 cm), this satisfies clinical applications without raising safety concerns. Hardware differences—higher gradient amplitude, increased receive channels, and new RF coils—do not compromise safety; peripheral nerve and cardiac stimulation are controlled per IEC 60601-2-33. Software enhancements (AI-based features like ACS, DeepRecon, EasyScan, t-ACS, AiCo, and motion correction algorithms) are validated independently with diverse demographic cohorts and pass acceptance criteria; they improve workflow and image quality without introducing new diagnostic risks. Clinical image review by U.S. board-certified radiologists confirmed diagnostic quality equivalence across all protocols and body parts.