Espree

This study involved MRI examinations performed from 2010 to 2017 available throughout the institutional PACS, designed to include a heterogeneous collection of MRI scanners and imaging studies. This incorporated both 1.5-Tand 3-T MR scanners, including Siemens Aera, Espree, Verio, Avanto, Skyra, and Trio Tim and GE Discovery and Signa Excite scanners. Multi-phasic contrast-enhanced T1-weighted breath-hold sequences from standard institutional liver MR imaging protocols were used with acquisition times of 12–18 s. Several different gadolinium-based contrast agents were used (dosed at 0.1 mmol/kg), including Dotarem (Guerbet), Gadavist (Bayer), Magnevist (Bayer), ProHance (Bracco Diagnostics), and Optimark (Covidien). Post-contrast images were analyzed, including late arterial phase (~20 s post-injection), portal venous phase(~70spost-injection),anddelayedvenousphase(~3min post-injection). Imaging parameters varied across different scanners and time frames; however, the majority were in the range of TR 3–5 ms, TE 1–2 ms, flip angle 9–13°, bandwidth 300– 500 Hz, slice thickness 3–4 mm, image matrix 256 × 132 to 320 ×216, and field-of-view 300 ×200 mm to 500 ×400 mm.

MRI examinations were conducted on 1.5T or 3T MRI scanners including Signa Excite^®, GE Discovery^®, Siemens Aera^®, Espree^®, Verio^®, Avanto^®, Skyra^®, and Trio Tim^® scanners. Non-contrast T1 images were acquired in all patients prior to administration of intravenous contrast. After the administration of intravenous gadolinium-based contrast agent (including Gadavist^® (Bayer), Dotarem^® (Guerbet), Magnevist^® (Bayer), ProHance^® (Bracco Diagnostics), and Optimark^® (Covidien), dosed at 0.1 mmol/kg), three T1-weighted three-dimensional (3D) gradient-echo (GRE) breath-hold imaging series (acquisition times of 12-18s, with fat suppression) were acquired reflecting CT/MRI LI-RADS recommendations: (1) late arterial, (2) portal venous and (3) delayed or equilibrium phase. Bolus tracking was applied in a large proportion of patients. Imaging parameters were in the range of TR 3-5 ms, TE 1-2 ms, flip angle 9-13 degrees, bandwidth 300-500 Hz, slice thickness 3-4 mm, image matrix 256 x 132 to 320 x 216, and field-of-view 300 x 200 mm to 500 x 400 mm. If a patient received multiple MRI scans, then the MRI performed closest to the date of pathological confirmation was used.