Lesion size and diffusion are the only quantitative parameters that are consistently measured in clinical liver imaging; radiological evaluation is otherwise almost completely qualitative. Dynamic contrast enhanced (DCE) imaging, which is the most critical set of acquisitions for liver, is seldom evaluated with quantitative measurements. Quantitative imaging methods attempt to evaluate focal lesions non-invasively and several ultra-fast free breathing perfusion MRI techniques have been developed (1–6). However, the experience with these techniques is often from single academic centers in Western countries, and wider multi-site experience in clinical practice worldwide is lacking. Here we describe free-breathing, 3D Through-time Spiral GRAPPA accelerated quantitative perfusion (7) in evaluating liver masses in a multi-institutional study in China and the US, at two different field strengths. This IRB-compliant, prospective study was conducted at 1.5T field strength in China (Siemens Aera, Erlangen, Germany) and 3T in the USA (Siemens Skyra, Erlangen, Germany). Patients with focal liver lesions who underwent a liver MRI exam as part of care at two tertiary care hospitals in China and USA were recruited. Contrast-enhanced T1-weighted 3D volumes were acquired using an interleaved variable-density stack-of-spirals spoiled gradient echo sequence with fat suppression and an in-plane acceleration factor of six (non-commercially available prototype). 100 volumes with a true temporal resolution of 1.6 to 1.9 seconds and a spatial resolution of 1.9x1.9x3.0 mm³ were acquired. After the acquisition of the 5th volume, 0.1 mmol/kg Gadobenate (Multihance, Bracco Diagnostics, Princeton, NJ) was injected at a rate of 3 mL/s followed by 20 mL of saline flush. The undersampled data were reconstructed off-line and images were registered to reduce the effects of respiratory motion and dynamic contrast changes among frames. Gd concentration-time curves in the aorta, portal vein and liver were generated and fit to a dual-input single compartment model (8) to derive arterial fraction (AF), distribution volume (DV) and mean transit time (MTT). The data acquired from the two groups were compared using a two-tailed Student’s t-test. A total of 50 lesions were evaluated in 32 patients from China and 18 patients from USA. A single representative lesion was evaluated in patients with multiple lesions. 28 hepatocellular carcinoma (HCC) and 18 metastases ranging in size from 0.5-8.5 cm were evaluated. Representative images are shown in Figures 2-4. The mean AF, DV and MTT for HCC and metastases are listed in Table 1. A statistically significant difference was noted in the MTT of metastases compared to HCC (p=7.07x10^-8). Four presumed benign lesions (2 focal nodular hyperplasia and 2 hemangiomas) were encountered but these were not biopsy proven and hence were insufficient for statistical analysis.High quality abdominal MRI exams are especially difficult to perform in a busy clinical practice where table time is constrained, particularly in countries such as China where the time pressure on scanners is greater than in the US. Many patients have difficulty holding their breath for the examination, and language barriers further complicate breath-hold instructions. Even if successfully completed, the enhancement characteristics of the lesions so detected frequently overlap. A free breathing quantitative perfusion MRI sequence overcomes all these limitations. We evaluated the robustness of quantitative perfusion measurements with 3D free breathing through-time spiral GRAPPA to evaluate liver lesions at large tertiary care hospitals in China and in the US, at two different field strengths. The acquisition was push-button and designed to not require “tweaking” by the technologist. Despite the geographical, temporal and language impediments to collaborating, successful quantitative imaging was possible. The AF, DV and MTT for HCC and metastases on both machines at two field strengths are in agreement with each other and with the values previously reported in literature (9-12). Both hepatocellular carcinoma and metastases derive a greater fraction of their blood supply from hepatic artery and hence have a high arterial fraction; the higher DV values are attributed to high cellularity of these lesions. The higher mean transit time for metastases is attributed to fibrosis and compression of microvasculature leading to increased tissue resistance (11).The work shows that quantitative perfusion parameters obtained from quantitative modeling of a 3D free breathing through time spiral GRAPPA acquisition are in agreement across institutions, continents and field strengths. In addition to the comfort provided to both the patient and the technologist by a free breathing exam, the high temporal resolution made possible by such an approach enables a quantitative dimension to the contrast enhanced liver exam.