Liver cirrhosis is an abnormal liver condition that the liver would repair it through the deposition of collagen, proteoglycans, and other macromolecules in the extracellular matrix [1]. The risk of liver cancer is greatly increased once cirrhosis develops. A non-invasive method that can objectively and simply assessment and grade the liver fibrosis is clinically required. T1ρ relaxation time has been proven to relevant with the macromolecular composition and proton exchange of tissues[2]. It may play as a non-invasive biomarker to investigate liver fibrosis. The first whole-liver study was carried out on 1.5T MR Scanner [3]. The purpose of our study was to implement the T1ρ method with whole-liver coverage that is breathing-hold free, and to initially apply the method to evaluate severity of whole-liver cirrhosis non-invasively at 3.0T.This prospective study was approved by the local ethics committee, and written informed consent was obtained from each participants. Twenty four healthy volunteers (Mean age 39.13 yrs. age range 23-63 yrs. 9F/15M) and thirty patients (Mean age 49.83 yrs. age range 28-75 yrs. 9F/21M) with clinically diagnosed different stage of liver cirrhosis were recruited for the experiments. The MR experiment was conducted on a Philips 3.0T clinical scanner (Achieva TX, Best, the Netherlands). A 16-channel SENSE Torso XL coil was used for signal reception. The specific scan parameters for whole-liver T1ρ sequence were: 3D T1-TFE, field of view = 400 x 352mm2, resolution=2.0 x 2.4 x 7.0mm3, duration of saturation pulse=1000ms, TR/TE = 5.1/2.5ms, flip angle=10°, spin lock frequency=500 Hz, spin lock times=0, 10, 20, 40, and 60ms, each spin lock scan time=2min56sec. The inter-reader agreement between the two independent radiologists in measuring the T1ρ values was assessed. Mean T1ρ values of the three groups (Normal liver, Child-Pugh A and Child-Pugh B&C) were compared by one-way analysis of variance . All participants were scanned successfully. The overall imaging quality was acceptably evaluated by two radiologists. Six to eight ROIs were placed on liver parenchyma, which excluded visible artifacts and large blood vessels in the image. The intraclass correlation coefficient of the interreader agreement was 0.989 which indicated an excellent interreader (Fig. 1). The mean T1ρ value (mean ± SD) of the healthy volunteers was 37.91 ± 3.37ms, (ranged from 31.83 to 43.67 ms). The mean T1ρ value of the patients (Child-Pugh A: 45.77±2.60, Child-Pugh B&C: 53.74±4.32) were significantly higher than those of the healthy volunteers (P<0.001). As shown in Fig. 2. An increase trend was also found as the Child-Pugh stage increase (Fig 3) .Respiratory trigger was used to minimize spatial misregistration of different spin lock times. Volume shimming was applied to minimize B0 inhomogeneity. In addition, a rectangular 180°refocus pulse was inserted into middle of the spin lock pulse to further reduce the banding artifacts causing by the B0 inhomogeneity. T1ρ relaxation time can reflect excess deposition of collagen or extracellular matrix macromolecules in liver, and our initial experiment showed a positive results for the feasibility of liver cirrhosis stage using quantitative 3D whole liver T1ρ mapping at 3.0T. The mean T1ρ of cirrhosis patients were significant higher than healthy volunteers, and the mean T1ρ were also increased as the Child-Pugh classification increased.Conclusions: We propose for the first time a 3D whole-liver coverage T1ρ sequence at 3.0T with reliable quantifications, because this method provides an alternative approach to non-invasive evaluation of the whole-liver cirrhosis. We propose for the first time a 3D whole-liver coverage T1ρ sequence at 3.0T with reliable quantifications, because this method provides an alternative approach to non-invasive evaluation of the whole-liver cirrhosis.