NI PCI Powers a SensibleTAB Arm Rehabilitation Robot

  Print Print

"SensibleTAB rehabilitation treatment increased the shoulder and arm function recovery in most hemiplegic stroke patients who had not previously obtained any improvement of motor functions, despite months of therapy using only conventional therapies. Without PCI-7342, LabVIEW, and NI Thailand staff support, this project would not have been successful."

- Parit Wongphaet, Samrong General Hospital

The Challenge:
Creating high-repetition, task-specific arm rehabilitation therapy under a concept of cognitive sensory motor training, which consists of different types of sensory retraining in a virtual environment such as a solid wall, spring, damper, and force field, for stroke patients.

The Solution:
Using a PCI-7342 controller in a SensibleTAB arm rehabilitation robot, we successfully controlled two-degrees-of-freedom servo motors using cognitive sensory motor training in an impedance control mode, which modulates how the robot reacts to mechanical perturbation caused by the patient.

Author(s):
Prakarnkiat Youngkong - Institute of Field Robotics, King Mongkut’s University of Technology Thonburi
Sarayut Hatsanai -
Parit Wongphaet - Samrong General Hospital

Overcoming Sensory Retraining Limitations

Stroke is one of the most common neurological causes of disability, and the number of patients has been increasing. It is widely perceived that high-repetition, task-specific, and impairment-appropriate training results in better recovery. In fact, sensory retraining yields better motor function recovery. In motor learning, feedback is essential in determining training efficiency. However, daily clinical practices provide limited practice repetition. It is simply too daunting for human therapists to deliver crucial, constant real-time feedback of each of the hundreds of repetitions during a session for multiple patients. Recently, Cochrane Collaboration review has shown that using robotic rehabilitation devices could result in better arm function recovery than conventional therapy. Existing robots—for instance, MIT-Manus from MIT and ARMin from ETH Zurich—are different in the way that they deliver training therapy to patients. But none of them are specifically designed for sensory re-education and interactive simulated virtual object palpation exercises. A group of researchers in the Mechanical Engineering Department at Thammasat University, the Institute of Field Robotics, and Samrong General Hospital in Thailand, designed and developed SensibleTAB, a novel robotic arm trainer illustrated in Figure 1. SensibleTAB could not only assist patients in training on a horizontal and tilted plane, but also perform sensory retraining and simulate interaction with virtual objects.

SensibleTAB System Setup

According to Figure 3, SensibleTAB is divided into two major components:

  1. The control tower (blue area)
  2. The X-Y table (red area), where therapists who conduct the training and stroke patients will interface, respectively.

The PCI-7342 is connected directly to the PC on which the rehabilitation program, which uses LabVIEW, is installed. We constructed the main servo motor control program in block-diagram style, as shown in Figure 4.

The therapist can select one of three modes:

  • Sensory Retraining
  • Virtual Object
  • Rehabilitation Games

from a touch screen attached to the control tower. We created the pleasant main therapy GUI using LabVIEW, as shown in Figure 5.

Sensory Retraining Mode

In sensory retraining mode, the therapist has to choose either point- or path-mode training for the patient. The big difference between these two modes is the way the patient senses the target.

In point mode, therapists create different point locations at the beginning of the session. Then, the patient passively familiarizes with each point location. Next, the patient will be instructed to close his or her eyes and start the training. The therapist will randomly select point locations and ask the patient on which point his or her hand is currently located. At the end, the related outcome displays and the test concludes, as illustrated in Figure 8.

Path mode is quite similar to point mode, except that the therapist creates a pattern (or path) at the beginning of the session. The therapist selects a random pattern and asks the patient which pattern his or her hand is currently drawing.

Virtual Object Mode

In virtual object mode, the therapist has to choose either static or dynamic training for the patient. The big difference between these two modes is the way the environment is created.

Static Mode

In static mode, the therapist creates different sizes and shapes (for example, square or circle) at the beginning of the session. Then, the patient passively familiarizes with each point location. Next, the patient will be instructed to close his or her eyes and start the training. The therapist will select random sizes and shapes and ask the patient which size and shape his or her hand is currently touching. At the end of the session, the related outcome displays and the test concludes, as shown in Figure 13.

Dynamic Mode

In dynamic mode, therapists create different force field, spring, and damper magnitudes at the beginning of the session. Then, the patient passively familiarizes with each point location. Next, the patient will be instructed to close his or her eyes and start the training. The therapist randomly selects a virtual environment one at a time and asks the patient which set of environments his or her hand is currently touching. At the end of the session, the related outcome displays and the test concludes, as shown in Figure 15.

Clinical Trial Outcome

We initiated a clinical trial for 12 chronic stroke patients, equally male and female, with persisting hemiplegic arm paresis. They had an average of 17.75 weeks after the brain insult, and received 11.9 therapy sessions within 9.7 weeks. Despite the length of time between stroke onset and the start of robotic training therapy and limited therapy time (average of 11.15 sessions within 9.75 weeks), nine out of 12 patients showed increased motor function in the paretic arm after the therapy. Five of the patients showed a higher score on the (short-form) Fugl-Meyer scale, and another four patients showed some improvement in the Fugl-Meyer score.

Conclusion

We designed and developed SensibleTAB, an arm rehabilitation robot for stroke patient sensory re-education and interactive simulated virtual object palpation exercises, using a PCI-7342 motion controller and LabVIEW. SensibleTAB rehabilitation treatment increased shoulder and arm function recovery in most hemiplegic stroke patients who had not previously obtained any improvement of motor functions, despite months of therapy using only conventional therapies. Without PCI-7342, LabVIEW, and NI Thailand staff support, this project would not have been successful.

Acknowledgement

This project was financially supported by TMGI Co., Ltd., Thailand. We are also grateful for the excellent medical support from Samrong General Hospital, Thailand.

Bookmark and Share


Explore the NI Developer Community

Discover and collaborate on the latest example code and tutorials with a worldwide community of engineers and scientists.

‌Check‌ out‌ the‌ NI‌ Community


Who is National Instruments?

National Instruments provides a graphical system design platform for test, control, and embedded design applications that is transforming the way engineers and scientists design, prototype, and deploy systems.

‌Learn‌ more‌ about‌ NI