目的：研究一种用于脑瘫儿童踝关节运动康复的机器人，并初步验证其在临床应用中的可行性。方法：设计研发一套踝关节康复机器人系统，并在北京大学第一医院康复医学科进行临床初步试用，通过比较6例脑瘫儿童训练前后跖屈肌痉挛和挛缩程度来初步验证该设备的临床可行性和即刻疗效。训练前后选取改良Tardieu量表和踝关节跖屈力矩来评价跖屈肌张力和软组织顺应性，所有受试者接受5组踝关节被动牵伸和静态拉伸训练，每组训练包含10个周期。采用SPSS 19.0软件对训练前后数据进行统计分析。结果：所有受试者训练后踝跖屈肌群Tardieu量表评测R1、R2角度均较训练前显著增加（腓肠肌PR1=0.003，PR2=0.029；比目鱼肌PR1=0.002，PR2=0.034），R1与R2差值训练前后差异无统计学意义（P值分别为0.067和0.045），踝关节各背屈角度下跖屈肌对抗力矩训练后较训练前显著降低（0°时P=0.001，10°时P=0.001，20°时P=0.014，30°时P=0.002）。结论：踝关节康复机器人系统可即刻显著改善脑瘫儿童踝关节跖屈肌挛缩和软组织顺应性，患儿耐受良好，接受度和配合度高，适宜在脑瘫儿童康复中应用，但仍需进一步行大样本随机对照研究以明确其疗效。

Objective： To propose a kind of robotic ankle-foot rehabilitation system for children with cerebral palsy and to preliminarily verify its feasibility in clinical application. Methods: A robot assisted ankle-foot rehabilitation system was specially designed and developed for children with cerebral palsy and a preliminary clinical study was conducted in Department of Rehabilitation Medicine, Peking University First Hospital. Modified Tardieu Scale and joint biomechanical properties (ankle plantar flexion resis-tance torque under different ankle dorsiflexion angles) were measured to analyze the muscle tone and soft tissue compliance of the ankle plantar flexors pre- and post-robotic training intervention. Six children with cerebral palsy (4 girls and 2 boys, mean age: 7 years) were recruited in this study. Each subject received 5 session robotic training and each session included 10-cycle passive stretching and static hold. SPSS 19.0 software was used for data statistical analysis. Results: Both R1 and R2 angles of Modified Tardieu Scale for ankle plantar flexors after training were significantly higher than those before the treatments (Gastrocnemius: PR1=0.003, PR2=0.029; Soleus: PR1=0.002, PR2=0.034). The difference between R2 and R1 was of no statistical difference before and after the training (P=0.067 and P=0.067, respectively). After training, the ankle plantar flexion resistance torque under different dorsi-flexion angles (0°, 10°, 20°, 30°) were significantly reduced than those before training (P=0.001, P=0.001, P=0.014, P=0.002, respectively). Conclusion: The robot assisted ankle-foot rehabilitation system can improve the contracture and soft tissue compliance of cerebral palsy children’s ankle plantar flexors. All the children in the study were well tolerated and interested with the training, easy to accept and cooperate with it. This device may be suitable for application in the rehabilitation of children with cerebral palsy. However, further randomized clinical trials with larger sample size are still needed to verify the long term efficacy of this device.