Objective To analyze the diagnostic efficacy of the fatty liver index （FLI） in the screening of metabolic dysfunction-associated steatotic liver disease （MASLD） Methods Prospectively， 109 patients diagnosed with MASLD who were admitted to our hospital between June 2023 and June 2025 were selected as the research subjects. They were classified into the diabetes group （n = 42） and the non-diabetes group （n = 67） according to the presence or absence of type 2 diabetes. Additionally， 109 patients without MASLD during the same period were chosen as the control group. The baseline data and Fatty Liver Index （FLI） of different groups were compared. Binary logistic regression analysis was conducted to identify the influencing factors， and the receiver operating characteristic curve （ROC） was employed to assess the screening efficacy. Results The proportions of smoking history， hypertension， and type 2 diabetes in the MASLD group were significantly higher compared to those in the non-MASLD group （P < 0.05）. The levels of systolic blood pressure， diastolic blood pressure， waist circumference， BMI， fasting plasma glucose， HbA_1c， fasting insulin， total cholesterol， triglycerides， hs-CRP， GGT， ALT， AST， TBIL， and FLI in the MASLD group were significantly higher than those in the non-MASLD group， whereas the levels of LDL-C， ALB， and PA were significantly lower than those in the non-MASLD group （P < 0.05）. The prevalence of hypertension in MASLD patients with type 2 diabetes was higher than that in the non-diabetic group （P < 0.05）， and the age， systolic blood pressure， waist circumference， BMI， fasting plasma glucose， HbA_1c， fasting insulin， triglycerides， DBIL， and FLI levels were significantly higher than those in the non-diabetic group （P < 0.05）. Multivariate analysis indicated that FLI， type 2 diabetes， HbA_1c， and hs-CRP were independent predictors for MASLD screening （P < 0.05）； in contrast， FLI， hypertension， HbA_1c， and FINS were independent predictors for MASLD screening in type 2 diabetes patients （P < 0.05）. ROC curve analysis demonstrated that in screening for MASLD， FLI had the best efficacy （AUC = 0.886）， followed by hs-CRP （AUC = 0.755）； the combined prediction model had better efficacy （AUC = 0.916）， with a cutoff value of 0.433， a sensitivity of 88.1%， and a specificity of 81.7%. In type 2 diabetes patients with MASLD， HbA_1c performed best （AUC = 0.802）， followed by FLI （AUC = 0.724）； the combined model had the best efficacy （AUC = 0.876）， with a cutoff value of 0.480， a sensitivity of 95.5%， and a specificity of 77.1%. Conclusions FLI stands out as the most effective single predictor for MASLD screening. Moreover， its efficacy is further enhanced when it is combined with type 2 diabetes， HbA_1c， and hs-CRP. Among patients with type 2 diabetes， the combination of FLI， hypertension， HbA_1c， and FINS yields the optimal screening results for MASLD. As a non-invasive indicator， FLI holds significant application value in the early screening of MASLD.