Objective:To explore the risk factors affecting the short-term prognosis in patients with acute kidney injury(AKI),and to construct a risk prediction model and verify it.Methods:A total of 228 patients with AKI were chosen and split into modeling group(n=182)and verification group(n=46).The risk factors were determined by single factor and binary Logistic regression analy-sis,and the nomogram model was developed.And its effectiveness was evaluated using the ROC,calibration,and DCA curves.Results:Poor short-term prognosis occurred in 45 patients(24.73％)in the modeling group and 11 patients(23.91％)in the validation group.Platelet,blood urea nitrogen,serum creatinine,serum potassium,renal replacement therapy and length of stay were independent risk factors of poor short-term prognosis in patients with AKI.The AUC of the nomogram model in the modeling group was 0.975(95％ CI:O.956～0.994),and the AUC of the validation group was 0.930(95％ CI:O.858～1.000).The calibration curve analysis showed that the model fitted well.The net income of each range of the threshold probability value of the DCA curve analysis was greater than 0.Conclusion:The nomogram model has a good predictive ability for the short-term poor prognosis of AKI patients and can provide reference for medical staff to evaluate the short-term poor prognosis of AKI.