Bone metastasis is a leading cause of morbidity and mortality for patients with metastatic prostate cancer. Early detection of bone metastasis plays a key role in the assessment of treatment methods and prognosis. Although currently bone scintigraphy serves as the standard method to detect bone metastasis, whole-body MRI has been shown to be more sensitive. However, research on homogeneous tumor entities by using MRI is still uncommon. This study aims to evaluate the diagnostic efficiency of readout-segmented whole body diffusion weighted imaging and bone scintigraphy using 99mTc-MDP (BS) in the detection of bone metastases for patients with newly diagnosed high-risk or biochemical recurrent prostate cancer.From June 2014 to December 2014, 63 consecutive patients with newly diagnosed high-risk or biochemically recurrent prostate cancer were recruited for this prospective study. All patients underwent both readout-segmented EPI [1] for whole body diffusion weighted imaging (RESOLVE WB-DWI) on a 3T MRI scanner (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany) and bone scintigraphy completed with target scan. RESOLVE WB-DWI used the following parameters: TR/TE = 7500/68ms, FOV = 430 x 340 mm2, scan matrix = 176 x 176, slice thickness = 5mm, b = 0, 800 s/mm2, diffusion mode = 3D diagonal, 4 or 5 steps (depending on the patient’s height) with coverage from head to pelvis, total scan time =~21 mins. The MRI target scans included T2WI (STIR), T1WI and contrast enhanced T1WI if suspected lesions were found. Both examinations were completed within 10 days after the patients were included in this study. Patient- and lesion-based analyses were each performed by two experienced radiologists. All patients received a follow up exam after 4-6 months with RESOLVE WB-DWI or BS which were considered as the best valuable comparator (BVC) to evaluate the WB-DWI and BS diagnosis for sensitivity, specificity and accuracy. The chi-square test was used for the significant differential analysis.Based on the BVC, 20 patients (32%) had bone metastasis. In patient-based analysis, RESOLVE WB-DWI showed 18 true-positive findings while BS had only 12 true-positive findings. Three patients were diagnosed with uncountable bone metastases. In these patients, DWI showed more bone metastases than BS. When using lesion-based analysis, these patients were analyzed separately. RESOLVE WB-DWI showed earlier detection of bone metastasis than BS, as demonstrated in figure 1. For the 27 bone metastasis lesions which could be verified, RESOLVE WB-DWI was able to show 25 true-positive lesions and BS showed 16 true-positive lesions. The sensitivity, specificity, and accuracy of WB-DWI and BS for detecting bone metastases based on patients were 90.00% and 60.00%, 97.50% and 80.00%, and 95.00% and 73.33%, respectively (p<0.05). The sensitivity, specificity, and the accuracy of WB-DWI and BS for detecting bone metastases based on lesions were 92.59% and 59.26%, 97.78% and 77.78%, 95.83% and 70.83% , respectively (p<0.05).RESOLVE whole-body DWI shows higher sensitivity, specificity and accuracy than BS in bone metastasis detection, and is capable of early detection and discovers more bone metastases lesions caused by prostate cancer.