Abstract: Objective To study the application of isokinetic muscle testing in identification of the faked paralysis to provide scientific data for establishing a standard system of muscle strength in forensic medicine identification. Methods Fifty-seven patients with bone fracture or nerve damage as damaged group and 128 normal subjects pretended paralysis as faked paralyzed group were included in this study. Isokinetic muscle testing was performed on bilateral knees of all subjects in the two groups. The peak torque （PT） and peak torque angle （PTA） were compared between both sides in each group. The features of torque-time graph of two groups were classified. Results In the damaged group, the differences of PT between two sides of flexors and extensors were statistically significant （P＜0.05）, while the differences of PTA were not statistically significant （P＞0.05）. In faked paralyzed group, the differences of PT and PTA between two sides of flexors and extensors were both statistically significant （P＜0.05）. The torque-time graph of damaged knee presented mostly as single lead peak, while torque-time graph of the faked paralyzed knee presented mostly as multiple peaks. Conclusion The feature of torque-time graph could be useful to identify the faked paralyzed extremities in forensic authentication.

Key words: forensic medicine, hemiplegia, isokinetic muscle tests