Objective To investigate the relationship between the morphology and elasticity of the levator ani muscle （LAM） and stress urinary incontinence （SUI）， and to develop a multimodal diagnostic model for SUI based on LAM morphology and elasticity parameters， while evaluating the diagnostic performance of this model. Methods From September 2020 to September 2022， 147 female patients with SUI from the Affiliated Hospital of North China University of Science and Technology were enrolled as the SUI group （case group）， while 144 women without SUI during the same period were selected as the non-SUI group （control group）. Transperineal ultrasonography was conducted to measure the anteroposterior diameter （LH-A1） and transverse diameter （LH-D1） of the levator hiatus at rest， the resting area of the levator hiatus （LA1）， as well as the anteroposterior diameter （LH-A2）， transverse diameter （LH-D2）， and area （LA2） of the levator hiatus during the maximum Valsalva maneuver. Additionally， ultrasonography was used to observe LAM injury （LA-MI） during pelvic muscle contraction. Shear wave elastography （SWE） was also performed transperineally to record the elastic modulus values of the puborectalis muscle at rest （E1） and during pelvic muscle contraction （E3）. The differences in ultrasound parameters between the two groups were compared， and a logistic regression model was constructed for multivariate analysis to establish a diagnostic model for SUI. The goodness of fit of the logistic regression model was assessed using the Hosmer-Lemeshow test. The diagnostic performance of individual indicators and the diagnostic model for SUI was evaluated using the receiver operating characteristic （ROC） curve. Finally， the clinical utility of the model was assessed using decision curve analysis. Results There were statistically significant differences in age， BMI， LH-A1， LH-D1， LA1， LH-A2， LH-D2， LA2， LA-MI， E1， and E3 between the two groups （P < 0.05）. Multivariate logistic regression analysis revealed that age， BMI， LH-A1， LA2， LA-MI， E1， and E3 were significantly associated with SUI （P < 0.05）. Based on these findings， a diagnostic model for SUI was established： PRESUI = 0.261 × age + 0.904 × BMI-4.300 × LH-A1 + 1.166 × LA2-2.815 × LA-MI + 0.587 × E1-0.631 × E3-1.258. The model demonstrated excellent goodness-of-fit （P = 0.983）. The ROC curve analysis indicated that age， BMI， LH-A1， LA2， LA-MI， E1， and E3 all exhibited diagnostic efficacy for SUI （AUC > 0.500， P < 0.05）. Notably， the AUC of the constructed diagnostic model for SUI was 0.996 （95%CI： 0.992 ~ 1.000）， suggesting that the diagnostic accuracy of the model surpassed that of individual indicators. When the cut-off value of the diagnostic model was set at 0.437， the sensitivity reached 98.0%， and the specificity was 97.2%. Furthermore， the decision curve analysis demonstrated that the diagnostic model provided substantial net clinical benefit within the threshold probability range of 0.1 to 1.0. Conclusions The morphology and elasticity of the LAM are significantly altered in women with SUI. The SWE technique demonstrates potential application value for quantitatively assessing the elasticity of the LAM. Furthermore， the diagnostic model constructed based on age， BMI， LH-A1， LA2， LA-MI， E1， and E3 exhibits high clinical application value.