uMR 780

The uMR 780 is a 3.0T magnetic resonance diagnostic device designed to generate comprehensive anatomical and functional images for diagnostic purposes, employing various cross-sectional and spectroscopic techniques. This updated version enhances imaging capabilities through software advancements, including novel pulse sequences, AI-driven compressed sensing, and sophisticated image processing methods, while retaining the original hardware and core MRI science, supported by non-clinical performance data.

The uMR 780 is a 3.0T superconducting magnetic resonance diagnostic device (MRDD) featuring a 65cm patient bore, designed to produce sagittal, transverse, coronal, and oblique cross-sectional images, as well as spectroscopic images. These images display internal anatomical structures and/or functions of the head, body, and extremities, aiding trained physicians in diagnosis; contrast agents can be utilized depending on the scan's region of interest. The system is an update to a previously existing device (K191157), maintaining the same hardware configuration but introducing significant software modifications. The modifications include the addition and alteration of pulse sequences, such as new sequences (gre_quick_wfi, hise, gre_quick_4dcemra, gre_ute, gre_maps), broadened contrast characteristics for T1, T2, and Pd sequences, and new options like dark blood, navigator, multi-echo, reduced-FOV, and computed DWI. Furthermore, new reconstruction methods like compressed sensing and AI-assisted compressed sensing have been incorporated, alongside additional imaging processing methods including FACT (Fat Analysis and Calculation Technique), PSIR (Phase Sensitive Inversion Recovery), cDWI (Computed DWI), Inline T1/T2* Map, and SWI+ (Susceptibility Weighted Imaging Plus). Non-clinical performance evaluations for Spectroscopy, Computed DWI, and AI-assisted compressed sensing demonstrated that the device performs as expected, without altering the fundamental MRI scientific technology. No new clinical testing was conducted for these modifications.

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 do the new pulse sequences (e.g., gre_quick_wfi, hise, gre_quick_4dcemra) specifically improve diagnostic capabilities or image quality compared to the predicate device's sequences?; What is the quantitative impact of AI-assisted compressed sensing on image acquisition time, signal-to-noise ratio, and artifact reduction in the uMR 780 across different anatomical regions?; How does the performance of the new imaging processing methods, such as FACT and Inline T1/T2* Map, compare to established gold-standard techniques for fat quantification and T1/T2 mapping respectively?; What specific clinical scenarios or patient populations are expected to benefit most from the broadened application scope of contrast characteristics (T1, T2, Pd) and added sequence options like dark blood or computed DWI?; Given that non-clinical testing focused on Spectroscopy, Computed DWI, and AI-assisted compressed sensing, what validation was performed for the other new pulse sequences and image processing methods like PSIR and SWI+?