Objective To evaluate the value of MRI imaging data and biomarker detection in evaluating the prognosis of patients with traumatic brain injury. Methods 76 patients with TBI diagnosed in our hospital from April 2018 to April 2019 were enrolled. The patients were divided into MRI examination, MRI negative group (n=30), traumatic bleeding group (TH group, n=26), and traumatic microvascular injury (TVI group, n=20). The levels of vWF, cFn, and pNFH in plasma were measured by enzymelinked immunosorbent assay (ELISA). GRE, FLAIR, DWI, and SWI positive results were scored based on the number, size, and location of lesions. The clinical outcome of the patients was assessed using the Glasgow outcome scale (GOS). Results The MRInegative group had a GOS score of ＞3, which was significantly higher than that of the TH group and the TVI group (P＜0.05). Less than 48 hours after injury, the level of vWF in the TVI group was significantly lower than that in the MRInegative group and the TH group (P＜0.05), and the pNFH level was significantly higher than that in the MRInegative group and the TH group (P＜0.05). More than 48 hours after injury, the level of vWF in the TH group was significantly higher than that in the MRInegative group and the TVI group (P＜0.05). The levels of cFn and pNFH in the TVI group were significantly higher than those in the MRInegative group and the TH group (P＜0.05). The logistic regression HosmerLemeshow goodnessoffit test showed that when the levels of vWF, cFn and pNFH measured at ＜48h after injury were added to the predictive model, the accuracy of predicting good prognosis increased from 5760% to 77 %~81%. Conclusion The levels of vWF and cFn in plasma of patients with TBI change significantly after injury. MRI imaging data and biomarker detection have a good evaluation value for the prognosis of patients with traumatic brain injury.