Magnetic resonance-guided high-intensity focused ultrasound (MRg HIFU) ablation is currently emerging nonsurgical options in the treatment of uterine fibroids, but poor therapeutic responses are shown on some fibroids with certain properties that are known to be resistant to HIFU heating. It’s necessarily to choose suitable fibroids prior to MRg HIFU treatment by mean of screening MR examinations[1]. When post-treatment, nonperfused volume (NPV) of fibroids calculated from the CE-T1WI and symptom severity score (SSS) of patients obtained from the Fibroid Symptoms quality Of Life Questionnaire are the important indexes to evaluate therapeutic effect of fibroids and symptom relief of patients, respectively[2,3]. 3D histogram of apparent diffusion coefficient (ADC) analyzes the ADC values of the entire volume of the tumor, which could provide quantitative information about the tissue characteristics and heterogeneity[4,5]. Therefore, the aim of the study is to investigate the variation among screening fibroids through analysis of ADC histogram, in order to predict fibroids residual NPV proportion (residual NPV%=NPV at 6 months follow up/ NPV immediately after treatment ) and patients SSS, and expect to acquire the quantitative indication for selecting suitable fibroids to MRgHIFU ablation.Thirty five patients with 35 fibroids were treated by HIFU ablation on a clinical MR-guided HIFU system (Sonalleve; Philips Medical Systems, Vantaa, Finland). All MR images were acquired with Philips 1.5T MR scanner (Achieva, Philips Medical Systems, Best, the Netherlands). The axial DWIs with b factors as 0 and 700 s/mm2 were obtained from screening fibroids. CE-T1WIs were scanned immediately after treatment and 6 months follow up. Quality of life outcomes, measured by the SSS were assessed at screening and 6 months follow up. The NPV were calculated at MR console immediately after treatment and 6 months follow up, respectively, and residual NPV% were obtained with Excel. ADCmean (mean ADC value of whole fibroid)、ADCq (the distance between the 25th percentile and the 75th percentile voxel in one lesion)、kurtosis (the degree of distribution peakedness) and skewness (a measure of distribution asymmetry) derived from ADC histogram were acquired using three softwares: ImageJ, Excel, SPSS. Patients were stratified into two groups based on the residual NPV%. Group 1 had a residual NPV% of 20% or greater (mean 39.70%, n=19), and group 2 had residual NPV% less than 20% (mean 8.64%, n=16). The differences of ADCmean,ADCq, kurtosis and skewness between two groups were tested by t-test. The differences of the SSS between screening and 6 months follow up were also compared by t-test, not only total patients but subdivided groups. Statistical differences with P<0.05 were considered significant.The quantitative histogram parameters (ADCmean, ADCq, kurtosis and skewness) and patients age of two groups were shown in Table 1 as mean± standard deviation. There were significant differences of ADCmean, ADCq, kurtosis between two groups (Fig 1). At 6 months follow up, the average SSS of all the patients in this study(23.37±10.71) was significant less than that at screening (31.23±14.17), p=0.011. In group 1, the average SSS of 6 months follow up (23.57±9.97) was also less than that of screening (33.11±12.92), p=0.015. But in group 2, there were not significant differences of the SSS between post-treatment(23.13±11.87) and screening(29.00±15.65), p=0.241(Fig 2).3D histogram analysis of ADC values can objectively assesses the tissue inherent characteristics and heterogeneity of the whole fibroid. Residual NPV% reveals the extent of myoma necrosis after thermal ablation, which is important index to assess the treatment response of MRg HIFU ablation. The bigger of residual NPV proportion, the better of ablation outcome, meanwhile the more extent of patient symptom relief. The lower ADCmean of group 1 revealed fibroids containing less smooth muscle cells, abundant collagen fibers, which was beneficial to the deposition of ultrasonic heat. It was one of the reasons of group 1 to achieve good residual NPV%. The average value of kurtosis in group 1 was -0.0352, which was very close to the Gauss distribution, revealing the uniformity in structure.Contrarily, the kurtosis in group 2 was significant high, and ADCq of group 2 was lower than that of group 1 accordingly, which reveled the heterogeneity of group 2 fibroids. It may be the reason of group 2 to obtain low residual NPV% and unobvious symptom relief. In sum, histogram analysis of ADC can provide the quantitative information to predict fibroids ablation outcome and patients symptom relief, which may be indicated as a useful screening tool to guide patients selection for MRgHIFU ablation.