Objective To study the diagnostic value of MRI for pulmonary embolism, so as to provide information for clinical decision. Methods The trials about MRI in the diagnosis of pulmonary embolism were searched in the following databases such as PubMed, EMbase, The Cochrane Library, CBM, CNKI and VIP. The data of the included trials were extracted, the methodological quality was evaluated in accordance with the quality assessment of diagnostic accuracy studies (QUADAS), and then meta-analysis was conducted using Meta-Disc 1.4 software. The weighted sensitivity and specificity were aggregated, as well as the summary receiver operating characteristic (SROC) curve. Further, the area under the curve (AUC) was calculated to evaluate the value of MRI in the diagnosis of pulmonary embolism. Results A total of 6 English articles involing 595 patients were included. The results of heterogeneity test revealed that there was statistical heterogeneity among the results of studies. According to the random effects model, the weighted sensitivity and specificity were 0.87 (95%CI 0.80 to 0.91), and 0.98 (95%CI 0.96 to 0.99), respectively; and the AUC was 0.988 7. Conclusion MRI is a better non-invasive method as a routine examination for pulmonary embolism. It shows fairly high sensitivity and specificity, and has a good clinical value.