BlogGeneralWearable elbow robot in rehabilitation after a stroke

Wearable elbow robot in rehabilitation after a stroke

Authors: Irina Benedek, Victor Dabala, Oana Vanta 

Keywords: stroke, wearable elbow robot, robotic neuro-rehab, early-stroke rehab

The burden of stroke

What is the role of wearable elbow robots in rehabilitation after a stroke? Acute stroke, also known as a cerebrovascular accident, represents the acute onset of a focal neurological deficit due to an impairment in a specific vascular territory. Annually, approximately 800.000 new cases are reported only in the United States; in other words, there is a new case every 40 seconds, and death occurs every 4 minutes. Therefore, stroke remains one of the most important causes of mortality and disability worldwide and a significant public health issue [1, 2].

There are two main types of stroke: ischemic stroke (see Figure 1 for Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification) -accounting for 85% of acute cerebrovascular events- and hemorrhagic. Ischemic stroke is due to the interruption of blood flow in a particular brain territory, while a rupture of a blood vessel supplying the brain leads to a hemorrhage [1]. Hemorrhagic strokes are broadly due to intracerebral hemorrhage or subarachnoid hemorrhage [2].

The categories of ischemic stroke are showcased in Figure 1 below  [2].

Graphic 1 Ischemic Stroke cathegories

Figure 1. Ischemic stroke categories

A vast etiology can lead to a cerebrovascular attack; the most common risk factors are presented in Figure 2  below.

Graph 2 Common risk factors for stroke

Figure 2. Common risk factors for stroke

When discussing the consequences of stroke, it is estimated that approximately 70% of the patients who survive a stroke will suffer from hemiparesis affecting the upper limbs, but most of the patients will remain with a type of neurological impairment. In addition, two-thirds of post-stroke patients will also experience a considerable impairment in upper limb motor function. As a result, stroke limits the ability to perform daily activities ( See Figure 3).

Graph 3 Daily Activities limited by stroke

Figure 3. Daily activities limited by stroke

Possibilities of recovery after a stroke

One of the main goals of stroke rehabilitation is to improve upper limb motor function to reduce disability and enhance the overall quality of life. The most crucial part of long-term recovery was reported in the first three months when spontaneous recovery and learning-dependent processes were dominant. However, scientific data suggests rehabilitation should not be limited only to this period. Unfortunately, several studies have demonstrated that hospitalized patients were inactive most of the day in the ward. The time estimated for physical and occupational therapy was limited to approximately three hours per day. Active and passive kinesiotherapy duration was about 30 minutes, while median repetition was 33-50 or even less for patients with severe deficits. Despite these aspects, it has been recognized that the most effective way to increase neuroplasticity and motor recovery after a cerebrovascular attack is through intensive treatments and repetitive motor tasks [3].

Learn more about:

Robots aid in physical therapy for motor recovery of the upper limb

Over the past years, rehabilitation robots have gained popularity in specific stroke rehabilitation since they have shown the capability to increase and individualize the number of movement repetitions compared to classic kinesiotherapy. These advantages can offer clinicians the possibility to adapt the recovery programs according to every patient’s needs and better chances of achieving the desired goals. Even though almost every patient with motor impairments should be considered for this type of rehabilitation, recent studies proved that chronic stroke patients with moderate deficits showed better improvements in motor function from robotic-assisted upper limb training than those with mild impairments. Therefore, the stratification of stroke patients based on their motor deficits was essential to obtain the best individual outcomes. Since the elbow joint is the most common and highly affected after a stroke, the 2021 study of Huang MZ et al. highlights the development of a portable exoskeleton robot targeting upper limb neurorehabilitation with a focus on flexion/extension of the elbow [3]. 

As such, the study was bidirectional, i.e., it was meant to:

  • Assess the effectiveness and suitability of a wearable elbow robotic device for in-bed training of patients with subacute stroke
  • Evaluate the patterns of the active motor recovery of the upper limb in patients with moderate and severe disabilities at this level [3].

The study included 11 patients (mean age 50 years) with early subacute stroke (7 days to 3 months) and late subacute stroke (3 to 6 months) who followed the eligibility criteria presented in Figure 4 below [3].

Graph 4 Patient eligibility

Figure 4. Eligibility criteria

The robotic device

The exoskeleton consisted of the upper arm and forearm braces, a servomotor, a gear head, and a computer screen for visual feedback. The wearable robot was designed to supply passive stretching, game-based active movement, and assessment of biomechanical properties such as muscle strength and elbow range of motion [3], as seen in Figure 5.

fnhum 15 669059 g001

Figure 5. A. The wearable elbow robotic device and B. Clinical in-bed setup (from Huang MZ et al. In-Bed Sensorimotor Rehabilitation in Early and Late Subacute Stroke Using a Wearable Elbow Robot: A Pilot Study [3]. Available here.

The functionality of the elbow robot

While lying on their back, patients wore the robot on the impaired upper limb with the shoulder at about 30 degrees flexion and 15-degree abduction. The device was mounted with one brace aligned to the elbow flexion axis and the output axis (Figure 5 – A), while the computer monitor was placed in front of the patient and adjusted as needed (Figure 5 – B) [3]. 

Each session consisted of passive stretching of the elbow (15 minutes), assisted active training through movement gameplay (15 minutes), and passive cool-down stretching (15 minutes). Elbow active range of motion (ROM) and maximum isometric voluntary contraction (MVC) of elbow flexors and extensor muscles were measured before and after each session. The training protocol was adapted to every patient’s needs. Patients benefited from four sessions per week during approximately a month of hospitalization, resulting in around 15 sessions [3]. 

Clinical outcome results such as upper limb motor recovery, spasticity, and muscle strength were followed in all patients before and after robotic training and 4 weeks after completion. In addition, parameters such as the Fugl-Meyer Assessment of the upper extremity, the Motricity Index, and the Modified Ashworth Scale were used to evaluate the progress [3].      

Some minor adverse effects were reported, such as:

  • Mild skin compression due to robot fixation
  • Muscle soreness after the first session of training (relieved after 24 h)
  • Mild fatigue after the active movements.

Nonetheless, significant increases in the Maximum Voluntary Contractions were observed after the completion of training, with each session producing training-induced changes in the MVC of elbow muscle groups (1.93 Nm (muscle torque) increase in elbow flexors, 0.68 Nm (muscle torque) increase in elbow extensors).

Regarding the flexor muscle group, stroke survivors with severe upper limb disability had a lower initial performance value than the subjects with moderate motor deficits. By comparison, the post-session improvement rate was more prominent in the group of patients with severe motor impairment [3].

Patients were assessed both before and after the 15 training sessions and at four weeks after the end of the sessions, with no differences reported when comparing the data between the end of the sessions and the follow-up at 4 weeks. Moreover, no significant change was observed in coordination and muscle spasticity (mild or absent in most patients) [3].

The advantages of robotic training 

Robotic-assisted rehabilitation has a significant advantage, as this type of intervention can deliver higher dosage and higher-intensity training than conventional therapy. This aspect makes robotic-assisted therapy a promising novel technology for the rehabilitation of patients with motor deficits caused by stroke. Exoskeleton devices used in robotic-assisted therapy were found to have better rehabilitation results than conventional physiotherapy, and their efficacy was found to increase further when combining conventional therapy with robotic training [4]. 

The in-bed rehabilitation robotic device and training developed by Huang et al. [3] was found feasible for both early and late subacute stroke survivors. The superior aspects of the robotic device in contrast to conventional therapy are presented in Figure 6 below.





Graph 5 Superior aspects of the in bed rehabilitation robot

Figure 6. The superior aspects of the in-bed rehabilitation robot


Their [3] findings highlighted that applying in-bed rehabilitation robotic training in the subacute stroke stage is beneficial for motor recovery and is a patient-centered procedure. 

Future of the robotic elbow device in rehabilitation 

Considering the presented study by Huang et al. [3], it should be acknowledged that the use of such robotic elbow devices in patients with increased upper limb spasticity will help with the implementation of different clinical variables to evaluate muscle tone, synergistic movements, and active isolated movement in everyday practice in a more significant number of stroke survivors while enhancing a detailed analysis of the interferences between conventional rehabilitation and robotic procedures. 

Moreover, further developments of the current applications of robot-aided interventions in post-stroke therapy are expected to be used increasingly as robot-aided rehabilitation programs become more feasible and help improve stroke survivors’ clinical measurements and quality of life.


  1. Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J et al. Classification of Subtype of Acute Ischemic Stroke Definitions for Use in a Multicenter Clinical Trial. Stroke Vol 24, No 1 Jan 1993, 35-41. doi/10.1161/01.STR.24.1.35
  2. Tadi P, Lui F. Acute Stroke. [Updated 2021 Sep 29]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2022 Jan-. Available from:
  3. Huang MZ, Yoon Y-S, Yang J, Yang C-Y, and Zhang L-Q. In-Bed Sensorimotor Rehabilitation in Early and Late Subacute Stroke Using a Wearable Elbow Robot: A Pilot Study. Front. Hum. Neurosci. 2021. 15:669059. doi: 10.3389/fnhum.2021.669059
  4. Chang WH, Kim YH. Robot-assisted Therapy in Stroke Rehabilitation. J Stroke. 2013;15(3):174-181. doi:10.5853/jos.2013.15.3.174

Leave a Reply

Your email address will not be published. Required fields are marked *

Contact: [email protected]

Administrative Office: Cluj-Napoca, Romania

EFNR is the main forum for all professionals interested in neurorehabilitation across Europe. 


© EFNR 2022. Developed by & Digital Strategy by BlogDigital