OBJECTIVE： To establish a risk scoring system for identifying invasive fungal infection（IFI）patients with hematologic diseases. METHODS： Risk factors were investigated among 200 patients diagnosed with IFI and 200 control patients at the same time from Jan. 2008 to Dec. 2015. The single factor analysis and Logistic multivariate regression analysis were conducted for potential risk factors to screen and assign risk factors of IFI. The risk scoring system of IFI was established，the performance of scoring system was evaluated by receiver operating characteristic （ROC） curve. Using the scoring system， 103 patients of validation group were scored during Jan. to Jun. in 2016. The incidence of IFI in each group was compared. 18 high-risk patients were intervened by the scoring system. RESULTS： Community acquired infection， the reduction of neutrophils， fungal infection history， corticosteroids and broad-spectrum antibiotics （Enzyme inhibitors， glycopeptides， quinolones， aminoglycosides and carbapenems） were risk factors of IFI （P＜0.001）， and the score of them were 17， 10， 39， 14， 14 according to the regression coefficients. IFI risk scoring system was divided into low， medium and high risk （scoring 0-30， 31-40， ≥41）， AUC of ROC curve was 0.916. The incidence of IFI in low-risk， medium-risk and high risk groups were 3.0％， 10.7％ and 62.5％， high-risk group was significantly higher than low and medium-risk groups （P＜0.05）. The incidence of IFI was 16.7％ in intervention group， there was statistical significance compared to high-risk group of validation group （P＜0.05）. CONCLUSIONS： This scoring system shows good ability to distinguish risk stratification. It can help clinicians identifying IFI high-risk groups and timely guiding timely intervention for patients.