Objective: This study aims to compare the difference and the change trend of Muscle Architecture Parameters (MAP) between spastic and normal muscle tone patients after stroke, and analysis the application and value of Muscle Architecture Parameters in evaluating spasticity after stroke. Methods: 41 stroke patients were divided into spastic group (26 cases), normal muscle tone control group (15 cases). Modified Ashworth Scale (MAS) was applied in both groups for assessing muscle tone of lower limbs (no influence, contralateral). Muscle architectural parameters of ultrasound measurement include muscle thickness, fascicle length and pennation angle. The difference of three muscle architectural parameters between the affected side and the contralateral side in spastic group and the difference of MAS and three muscle architectural parameters between spastic group and normal control group were compared. Results: MAS score, Pennation Angle (PA) and Muscular Thickness (MT) value of MAP in spastic group were significantly higher than the control group, Fascicle length (FL) is significantly lower than the control group (P < 0.05). In spastic group, MAS score, PA and MT value of MAP of affected side muscle was substantially higher than that of contralateral, FL value significantly lower than that of contralateral (P < 0.05). There was positive correlation between MAS and PA and MT but was negative correlation between MAS and FL, rank correlation coefficient test was statistical significant (p < 0.05). Logistic multivariate regression analysis showed that spasticity can lead PA and FL to change (p < 0.05), there is no clear correlation between MT and spasticity occurs (P > 0.05). Conclusion: MAP has a better sensitivity in evaluating muscle tone between spastic patients and non-spastic patients, and degrees of spasticity have a clear corresponding exponential relationship to MAP. Combing MAS and MAP can assess muscle tone more objectively and accurately because subtle changes can be observed by testing values of architecture parameters that compensating for the shortcomings of MAS in reliability and validity. Thus it is helpful for guiding clinical antispastic practice.