GET THE APP

International Journal of Physical Medicine & Rehabilitation

International Journal of Physical Medicine & Rehabilitation
Open Access

ISSN: 2329-9096

+44 1300 500008

Mini Review - (2022)Volume 10, Issue 1

The Effectiveness of Rehabilitation Robotics Using Combined Sensors

Keisuke Kubota1* and Toshiaki Tsuji2
 
*Correspondence: Keisuke Kubota, Graduate Course of Health and Social Services, Saitama Prefectural University, Saitama, Japan, Tel: 818043340528, Email:

Author info »

About the Study

Robotics system using sensors may be categorized as passive mechanisms, Robotic manipulanda or Robotic exoskeletons. Many studies have demonstrated the effectiveness of rehabilitation using these sensor-based robots. In recent years, various computer technologies have worked together in collaboration in order to obtain better results. The development of sensor-based rehabilitation using machine learning technology can be useful for the accurate evaluation and therapeutic intervention in sensor-based rehabilitation. Thus, sensor-based robotics is evolving into highly accurate and more effective devices. Sensor-based robots could provide better information for building evidence in the rehabilitation field.

Robotics systems of rehabilitation

Robotics system using sensors may be categorized as passive mechanisms, Robotic manipulanda or Robotic exoskeletons [1]. Some of the devices are introduced below:

Passive mechanism

The passive mechanism is fitted with a range of sensors and often augmented with a virtual environment in the form of a game-like challenge to guide the patients through the training phase [1]. Through instrumentation such as built-in position sensors and software, the Armeo Spring can be engaged as an input device for the accomplishment of meaningful function tasks that are simulated in a virtual learning environment on a computer screen, with the provision of auditory and visual performance feedback during and after practice [2].

Robotic manipulanda

The robot manipulanda controls the motion at the end-effector of the robot. These systems have a variety of sensors for the measurement of hand motion and/or interaction forces [1]. A typical end-effector type device is the MIT-MANUS. The MITMANUS can move, assist or perturb the movement of a subject’s or patient’s upper limb and can record motion and mechanical quantities such as the position, velocity, and force applied [3].

Robotic exoskeletons

These are powered robotic mechanisms that regulate force and motion at the joint space of the subject’s limbs [1]. The Locomat is an exoskeleton robot combined with a treadmill and a trunk suspension system to become a body-weight supported passive gait robot [4]. The robot does not require voluntary movement by the patients but can achieve lower limb joint movement during gait with actuators and a treadmill. The ARMin is an upper limb exoskeleton robot that aids the patients to enforce their own movement trajectory for reaching movements, while the robot supports their efforts in order to complete the task [5].

Other systems

Sensor-based rehabilitation is not limited to systems that are integrated with robots. IMU can be attached to various parts of the body to acquire acceleration information. Nuesch, et al. have reported the RehaGait system, which includes seven IMU and joint angle calculations software, is similar in calculation accuracy to a motion capture system [6]. Moreover, IMU has been applied in combination with other devices. Milosevic, et al. proposed a system of motion tracking systems at home which combined IMU and video-based motion capture systems [7]. Wittmann, et al. developed a system using a therapy game and its assessment-driven target selection algorithm are implemented on the ArmeoSenso systems, a novel; IMU based virtual reality system for the training of upper limb function [8]. Many studies have demonstrated the effectiveness of rehabilitation using these sensor-based robots. On the other hand, effectiveness is limited and depends on the patient’s symptoms, amount of training and type of treatments. Therefore, caution is needed in interpreting the results which has also been reported [9]. In recent years, various computer technologies have worked together in collaboration in order to obtain better results.

The combination of machine learning

Rehabilitation using sensors has shown greater effectiveness in precise measurements and more objective assessment of the patient’s progress. Moreover, the quantified sensor information can be connected to the internet to autonomously collect evaluation results [10]. Creating a database of patient’s rehabilitation information affords numerous benefits. For example, by creating a database that contains information on various physical functions of patients, similar cases can be searched from the database. Tsuji, et al. proposed a motion matching technique for searching similar cases, which uses position and force information as a feature value [11]. The development of such database-based motion matching technology can be useful for the accurate evaluation and therapeutic intervention in sensor-based rehabilitation. Furthermore, quantified sensor information is combined with machine learning technology. Sekiya, et al. showed that the endpoint of the lower limb using end-effector type robot with a force sensor can estimate lower limb muscle strength with high accuracy from force date obtained from the endpoint of the lower limb [12]. Also, Wang, et al. proposed an algorithm using IMU combined machine learning techniques [13]. This system can estimate kinetic data during gait at knee osteoarthritis patients with data collected from IMU. Essentially, it is necessary to use a motion capture system and force plate to calculate the kinetic data during dynamic tasks but by using machine learning, it could be estimated from the IMU data. In this way, by simultaneously measuring multiple parameters in advance, it is possible to estimate various latent information from only one parameter.

Conclusion

Rehabilitation using sensor-based devices can provide accurate information for patient treatment effectiveness. Also, this technology enables highly accurate autonomous rehabilitation in a variety of situations. Furthermore, recently, with the development of computer technology, sensor information is being combined with machine learning to appropriately handle huge amounts of patient data. Thus, sensor-based robotics is evolving into highly accurate and more effective devices. Sensorbased robots could provide better information for building evidence in the rehabilitation field.

References

Author Info

Keisuke Kubota1* and Toshiaki Tsuji2
 
1Graduate Course of Health and Social Services, Saitama Prefectural University, Saitama, Japan
2Graduate School of Science, Saitama University, Saitama, Japan
 

Citation: Kubota K, Tsuji T (2022) Effectiveness of Rehabilitation Robotics using Combined Sensors. Int J Phys Med Rehabil. 10:621.

Received: 21-Dec-2021, Manuscript No. JPMR-22-12701; Editor assigned: 24-Dec-2021, Pre QC No. JPMR-22- 12701 (PQ); Reviewed: 07-Jan-2022, QC No. JPMR-22-12701; Revised: 11-Jan-2022, Manuscript No. JPMR-22- 12701(R); Published: 18-Jan-2022 , DOI: 10.35248/2329-9096-22.10.621

Copyright: © 2022 Kubota K, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Top