Wearable devices based on inertial measurement units through wireless sensor networks have many applications such as real-time motion monitoring and functional outcome assessment of stroke rehabilitation. However, additional investigations are warranted to validate their clinical value, particularly in detecting the synergy patterns of movements after stroke.To explore the feasibility and efficacy of wearable devices for upper limb rehabilitation in patients with chronic stroke and to compare the intervention effects (e.g., neurological recovery, active range of motion, and deviation angle) with those in a control group.A single-blind, randomized-controlled pilot study.Rehabilitation ward.A total of 18 patients with chronic stroke were randomly distributed into a device group and control group. Both groups received conventional rehabilitation; nevertheless, the device group was additionally subjected to 15 daily sessions at least three times a week for 5 weeks. The outcome measures included the upper extremity subscores of the Fugl-Meyer assessment, active range of motion, and deviation angle. These measurements were performed pre- and post-treatment.All five Fugl-Meyer assessment subscores improved in both the device and control groups after intervention; in particular, the “shoulder/elbow/forearm” subscore (p = 0.02, 0.03) and “total score” (p = 0.03, 0.03) substantially improved. The active range of motion of shoulder flexion and abduction substantially improved at pre-post treatment in both the device (p = 0.02, 0.03) and control (p = 0.02, 0.03) groups. The deviation angle of shoulder external rotation during shoulder abduction substantially improved in the device group (p = 0.02), but not in the control group.The designed wearable devices are practical and efficient for use in chronic patients with stroke.Wearable devices are expected to be useful for future internet-of-things rehabilitation clinical trials at home and in long-term care institutions.