Biomedical Research

Abstract

This brief review addresses the existing systems and challenges and provides future recommendations on computer- and biosensor-assisted rehabilitation systems for physically disabled patients. We further list the types of sensors, technical issues, and different software and hardware technologies that are currently used in rehabilitation systems to make the whole process dynamic and real-time. The review focused on 36 consolidated studies that were found using the following keywords: rehabilitation system, sensor, and computer. The electronic databases PubMed, Scopus, and Google Scholar were searched for relevant articles that were published from 2007 through 2012. These published articles included discussion of several biosensors, automated rehabilitation systems, and the application of these systems in the affected body parts of the individuals. We found that 54 types of biosensors have been used for real-time and computer-assisted rehabilitation systems. The findings suggest that there are still some body parts (such as the muscle tendons area and abdomen) and application areas (e.g., post- and pre-pregnancy) that have not yet been targeted by biosensor-supported medical rehabilitation systems aided by suitable hardware, software, and other assistive technologies.

Keywords

Biosensor, Signal, Computer, Hardware, Rehabilitation, Software, Disable Patients.

Introduction

Recent years have seen the advances of a number of automated and semi-automated systems to support physiotherapy and rehabilitation [1]. Automatic rehabilitation is a type of therapy that aims to partially or totally recover the neurocognitive function and motor abilities of a patient. Thus, the affected person needs a perfect machine- controlled recovery system that is easy to operate, can provide the correct result without delay, and is cost effective for the users. As a result, the whole process can help the patients return to their previous stage promptly. Due to the fast advances in medical technology worldwide, rehabilitation systems are currently used by patients after a major operation, chronic pain, sensory loss, stroke, unpredicted pain, severe accident, orthopaedic anarchy, brain injury, Parkinson’s disease (neurological disorder), psychological disorder, and sports-related injury and by older individuals [2,3]. Briefly, the cardiopulmonary, neurological, orthopaedic, paediatric, and integumentary (skin and related organ) systems are the most essential and common areas in the human body that are subjected to rehabilitation treatment [4]. The full recovery process may be accomplished after several days of daily rehabilitation, and the duration of the program depends on the patient’s health [5]. To monitor their progression, end users need vast and deep experiences with the rehabilitation devices that can aid a patient’s quick recovery.

A system for biosignal based device applications contains a biosignal sensor (e.g. biosensor) or multiple biosensors (e.g. multiple biosensors for EEG detection, or multiple biosensors that can each detect different types of biosignals), a biosignal processing unit, the apparatus (hardware), and a software application(s) that operate the biosignal information for different applications [6]. Biosensors are analytical devices that have two components: one is a bioreceptor and the other is a transducer. The bioreceptor is a biomolecule that recognises the target analyte, and the transducer converts the recognition event into a measurable signal [7]. In short, a biosensor converts a biological response into an electrical signal. There are several supported mechanisms that are used to actively monitor the rehabilitation process in real time. Physio logical sensors are one of the fundamental parts of biosignal processing, and are thus important in automatic rehabilitation systems because they offer considerable advantages in the clinical and biomedical industry, such as specificity, small size, faster response, and low cost [8].

Until now, several authors have discussed, developed and reviewed rehabilitation systems with different procedures. For example, Silvia et al., studied a novel mechatronicdependent neurorehabilitation system that can evaluate the post-stroke functional recovery through whole-body isometric force measurements [9]. Majdalawieh et al., explained the relationship between biomedical signals and rehabilitation engineering using EMG, EEG, neural network, wavelet transforms, and Fourier transforms [10]. In other studies, some scholars have reviewed home-based telerehabilitation systems for stroke patients. These studies focused on a few number of biosensors and applications of these biosensors in physical recovery systems and compared with the results obtained with the different sensors/ applications. Pantelopoulos et al. studied the uses of wearable biosensors for health monitoring systems [11,12]. Katherine et al., reviewed the clinical aspects of robotic assistive technology and human-robot interaction for rehabilitation but did not clearly mention the biosensors used in the study [13]. However, we did not find any previously reported work in which the authors summarised the biosensors, computer, hardware, and software used in various rehabilitation systems and the applicability of the different rehabilitation devises.

This review paper is organised as follows. In the Methods section, details on the overall searching process that was used to obtain the results are discussed, including the different keywords, the online databases, and the time frame. The biosensors, hardware, software, and computer process models that were found in the literature are summarised and presented briefly in the Results section through the use of a table. The Discussion section presents an overall discussion in which the new finding related to the development of new biosensors that may aid recovery systems are presented. The Conclusion section recaps the review paper.

Methods

A systematic search of the existing literature was performed using a combination of the keywords “rehabilitation system”, “sensor”, and “computer” to find the related studies published between 2007 and 2012 in the following electronic databases: PubMed, Scopus, and Google Scholar. During the article search, the relevant titles and abstracts were analysed. If the expected criterion was matched, the full text was then reviewed. The inclusion criteria were the following: (1) any rehabilitation and/or assistive system that is assisted by a biosensor and computer, (2) systems developed for the human body, and (3) papers written in English. The exclusion criteria were the following: (1) rehabilitation systems assisted by a robot or other devices (except a computer, PDA, smart phone, mobile etc.), (2) no information on the biosensor/ biosignal, and (3) not a real-time system for monitoring the signals.

Results

Literature search results

The three electronic databases contained 2,140 articles with the keywords “rehabilitation system”, “sensor” and “computer”. After the articles from the three databases were collected, all of the duplicate articles were removed to obtain 1,200 articles. All of the articles were then filtered to determine whether the systems are automatic, real-time, and assisted by a computer and biosensor, which resulted in a total of 165 articles. Then, 129 articles were rejected because these did not fulfil the review criteria. As a result, 36 articles went used in the analysis.

Rehabilitation systems

The 36 articles in which the researchers developed computerised rehabilitation systems are presented in Table 1. Of these systems, six were developed for different muscles [14-19], five were developed for cardiac patients [8,20-23], thirteen were developed for stroke patients [24-36], eight were developed for upper limbs [5,37-43], two were developed for chronic illness and neurolocomotor [44,45], one was developed for hemiplegia [46], and one was developed for ankle and spinal cord injury [47].

Table 1: An overview of biosensor-assisted automatic rehabilitation systems

Applied Sensors

The wide range of study designs included and reflected in this review paper incorporated various types of biosensors. Researchers mainly used electromechanical, electrical, optical, and thermal biosensors, acoustic signal transducers, or mass-sensitive biosensors to develop rehabilitation systems. In this systematic review, a total of 54 types of biosensors were used for rehabilitation systems; these are itemised in Table 1. Some of the most commonly used biosensors that were applied in rehabilitation systems are electromyography (EMG), galvanic skin response (GSR) Electrocardiography (ECG), electroencephalography (EEG), grip sensors, inertial measurement units, orientation, and torque. In addition, most of the developed systems used EMG for musculoskeletal rehabilitation. However, all the biosensors and their applications are listed in Table 1. Figure 1 represents a graphical view of the EEG signal recording of a child brain.

Figure 1: EEG signal recording process [48]

Software and hardware

Computer-assisted rehabilitation systems are significantly less labour-intensive compared with conventional manually assisted-movement therapy systems [4]. We discovered that a total of 36 computer-assisted rehabilitation systems have been developed and are described in the literature.

In these systems, the researchers indicated the various types of hardware and software information that are used as fundamental elements and bridges of communication for the rehabilitation technology. Some of the common software programs used in the rehabilitation systems were MatLab [14,15,24,36], LabVIEW [14], Visual C++[43], Java [23], a GSM-based electrical control system for a cell phone application during rehabilitation [20], virtual reality software [31,39], image processing software [38], and some open GL techniques. Additionally, some researchers used databases, such as PHP, MySQL, and Microsoft DB. Three studies did not name the software used in the rehabilitation system developed. In contrast, a number of hardware was used during the development of the described rehabilitation systems (Table 1). As shown in Table 1, researchers commonly utilised different types of microcontroller chips, computers, and other devices.

Both wireless and wired systems were preferred for signal recording.

Discussions

Assistive technology devices and accurate rehabilitation systems for individuals with motor disabilities caused by the aging process or stroke need a good human-friendly actuator that is compact and environmentally beneficial [49]. The review attempted to answer the following question: “which sensors are used in combination with various software programs, hardware, and other supporting methods to develop an automated rehabilitation system for an impaired human body?” This review attempted to gather the maximum number of software, hardware, sensors, algorithm, prototype, and framework options that are used in rehabilitation systems. Additionally, some restrictions in the rehabilitation mechanism were found. The key finding of this study are the following: 1) a total of 36 articles were reviewed, 2) 54 different types of biosensors have been used in recovery systems, 3) of all of the sensors, EMG and ECG were used more frequently for rehabilitation system development (Table 1), 4) the applicable areas of rehabilitation systems are the brain, heart, ankle, shoulder, finger, biceps, triceps, stroke, cardiac, post stroke, stress, eye, whole body monitor, chronic pain, and older individual (most of the systems were developed for the rehabilitation of the upper limb and stroke and cardiac patients), 5) subjects prefer to feel comfortable with noninvasive electrode placement than needle-based systems, 6) all of the rehabilitation systems provided assistance with a desktop computer or personal digital assistance (PDA), 7) modularity, portability, cost, compactness, and user friendliness are the main concerns of rehabilitation systems, and 8) a number of hardware and software are used to engineer real-time and online systems, e.g., different programming languages, such as C, Visual C++, Java, MatLab, LabVIEW, and web programming tools, microcontroller chips, operating systems, DAQ techniques, application of databases, 3-D models, web technologies, different PDA systems, and client-server database technologies.

The authors of this brief review are confident that recommendations should be based on strong and clear evidence. A powerful level of proof, which is based on reliable findings from a number of high-quality studies, is required, as described in this goal. Thus, based on the evidence, we make the following recommendations, 1) to date, some parts of the body have not been targeted for biosensor-assisted rehabilitation systems because these are out of the sight of the human body, e.g., inner muscles of the human body, muscle tendon area, abdomen, and pre- and post-pregnancy, 2) the advantages in the development of rehabilitation systems have been presented to facilitate future researchers on this endeavour, 3) the existing software and hardware technologies discussed are useful if combined to build a proposed computerised rehabilitation mechanism, and 4) feedback in real time is also important because a physically inactive person (and the therapist) expect clear and live data without delay, and 6) the aim of brief reviews should be to guide rehabilitation researchers toward the most effective and latest techniques. Moreover, this paper suggests that the following issues should be taken into account during the developing of a biosensor-assisted automatic rehabilitation system; first and foremost: online process, non-invasive sensor, affordable cost, various automated signal processing moralisations, not cumbersome, and interactive GUI. Secondary: Portable, i.e., small size, weight and space. There are a number of important limitations to this review. We do not include the robot supported rehabilitations systems or systems without any sensors. Consequently, the present review is unable to explore the availability of the the existing systems on the market (i.e. commercialized rehabilitation systems).

Conclusion

This reviews uncovered 36 papers that matched the defined search criteria: rehabilitation systems with various sensors, software, and hardware. The review also highlights that inexpensive, real-time, non-invasive, effective, easy-to-use, and portable systems are preferred by the end user. The authors of this review hope that this paper provides researchers a good understanding of biosensorassisted rehabilitation systems and their analysis procedures. This knowledge will assist them in the development of more powerful, flexible, real-time, and wellorganised applications related to recovery systems.

References

1. Bradley D, Acosta-Marquez C, Hawley M, et al. NeXOS – The design, development and evaluation of a rehabilitation system for the lower limbs. Mechatronics. 2009; 19(2): 247-257.
2. Ito S, Kawasaki H, Ishigure Y, Natsume M, Mouri T, Nishimoto Y. A design of fine motion assist equipment for disabled hand in robotic rehabilitation system. Journal of the Franklin Institute. 2011;348(1):79-89.
3. Gazzani F, Fadda A, Torre M, Macellari V. WARD: a pneumatic system for body weight relief in gait rehabilitation. IEEE Transactions on Rehabilitation Engineering 2000;8(4): 506-513.
4. Ahamed NU, Sundaraj K, Ahmad RB, et al. Recent Survey of Automated Rehabilitation Systems Using EMG Biosensors. J Phys Ther Sci. 2011; 23(6):945-8.
5. Mohamaddan S, Komeda T. Wire-driven mechanism for finger rehabilitation device. Int conf on mechatronics and automation, 2010; 1015-1018.
6. Alexander Z, Chuang C-i, Liu J, Westendorf D, Lee K. Biosignal based mobile device applications. US Patent 20,120, 295, 589; 2012.
7. Lee YH, Mutharasan R. Chapter 6 - Biosensors. In: Jon SW, editor. Sensor Technology Handbook. Burlington: Newnes; 2005. pp 161-180.
8. Gupta R, Bera JN, Mitra M. Development of an embedded system and MATLAB-based GUI for online acquisition and analysis of ECG signal. Measurement. 2010; 43(9): 1119-1126.
9. Petroni S, Mazzoleni S, Bellelli S, Cannizzo S, Palla I, Labella B, et al. Early assessment of neurorehabilitation technology: a case study. Int J Biomed Eng and Tech. 2010; 4(3): 232-244.
10. Majdalawieh O, Gu J, Bai T, Cheng G. Biomedical signal processing and rehabilitation engineering: a review. IEEE Conf on Communications, Computers and signal Processing 2003, 2: 1004-1007.
11. Pantelopoulos A, Bourbakis N. A survey on wearable biosensor systems for health monitoring. 30^th IEEE Int Conf on Engineering in Medicine and Biology Society 2008; 4887-4890.
12. Pantelopoulos A, Bourbakis NG. A Survey on Wearable Sensor-Based Systems for Health Monitoring and Prognosis. IEEE Trans. Syst. Man Cybern. Part CAppl Rev 2010; 40(1): 1-12.
13. Tsui KM, Yanco HA, Feil-Seifer DJ, Matari MJ. Survey of domain-specific performance measures in assistive robotic technology. Proc of the 8th Workshop on Performance Metrics for Intelligent Systems; Maryland, USA, ACM; 2008. pp 116-123.
14. Burns A, Greene BR, McGrath MJ, O'Shea et al. SHIMMERTM: A Wireless Sensor Platform for Noninvasive Biomedical Research. IEEE Sens J 2010; 10(9): 1527-1534.
15. Enzo Pasquale S, Gemignani A, Paradiso R, Taccini N, et al. Performance evaluation of sensing fabrics for monitoring physiological and biomechanical variables. IEEE T Inf Technol B. 2005; 9(3): 345-352.
16. Burdea GC, Cioi D, Martin J, Fensterheim D, et al. The Rutgers Arm II Rehabilitation System-A Feasibility Study. IEEE Transactions on Neural Systems and Rehabilitation Engineering 2010; 18(5): 505-514.
17. Bani Amer MM, Qtait Y, Al-Ebbini L, Ammary S, Awwad M. Design of a user-friendly LabVIEW-based toolbox for real-time monitoring and diagnosis of vital signals. Int J Med Eng Inform 2010; 2(3): 307-318.
18. Wu C-F, Liou J-J. PC-Based Rehabilitation System with Biofeedback. In: Karsh B-T, Ergonomics and Health Aspects of Work with Computers: Springer Berlin Heidelberg 2009; pp 204-211.
19. Rotariu C, Costin H, Arotaritei D, Constantinescu G. A Low Power Wireless Personal Area Network for Telemedicine. 4th European Con. of the Int Federation for Medical and Biological Engineering: Springer Berlin Heidelberg 2009; pp 982-985.
20. Worringham C, Rojek A, Stewart I. Development and Feasibility of a Smartphone, ECG and GPS Based System for Remotely Monitoring Exercise in Cardiac Rehabilitation. PLoS ONE 2011; 6(2): e14669.
21. Cartaya ME, Valles J, Meissimilly G, Botana G, Guerra A. MOVISHOW: Telemetry Software to Process ECG Signals of Multiple Patients in Rehabilitation. World Congress on Medical Physics and Biomedical Engineering 2009, Germany, pp 190-193.
22. Mattila J, Hang D, Mattila E, Sarela A. Mobile tools for home-based cardiac rehabilitation based on heart rate and movement activity analysis. IEEE Int Conf on Engineering in Medicine and Biology Society 2009; 6448-6452.
23. Sa, x, rela, A., Salminen J, et al. A home-based care model for outpatient cardiac rehabilitation based on mobile technologies. 3rd Int Conf on Pervasive Computing Technologies for Healthcare 2009: pp1-18.
24. Schabowsky CN, Godfrey SB, et al.,. Development and pilot testing of HEXORR: hand Exoskeleton rehabilitation robot. J Neuroeng Rehabil. 2010; 7: 36.
25. Xing S, Zhang X, editors. EMG-driven computer game for post-stroke rehabilitation. IEEE Conference on Robotics Automation and Mechatronics 2010; pp32-36.
26. Cameirao MS, Badia SB, et al. Neurorehabilitation using the virtual reality based Rehabilitation Gaming System: methodology, design, psychometrics, usability and validation. J Neuroeng Rehabil 2010; 7: 48.
27. Chee KL, Chen IM, Zhiqiang L, Song HY. A low cost wearable wireless sensing system for upper limb home rehabilitation. IEEE Conf on Robotics Automation and Mechatronics 2010, pp1-18.
28. Chuanchu W, Kok Soon P, Kai Keng A, et al. A feasibility study of non-invasive motor-imagery BCI-based robotic rehabilitation for Stroke patients. 4th Int IEEE Conf on Neural Engineering 2009, pp 271-274.
29. Novak D, Ziherl J, Olens, et al. Psychophysiological responses to robotic rehabilitation tasks in stroke. IEEE Trans Neural Syst Rehabil Eng 2010; 18(4): 351-361.
30. Kanna S, Heng J. Quantitative EEG parameters for monitoring and biofeedback during rehabilitation after stroke. IEEE Int Conf on Advanced Intelligent Mechatronics 2009 pp 1689-1694.
31. Steinisch M, Guarnieri BM, et al. Virtual reality and robotics for neuro-motor rehabilitation of ischemic stroke patients. World congress on medical physics and biomedical engineering2009,Germany pp 61-63.
32. Sasidhar S, Panda SK, Jianxin X. Design of a myoelectric glove for upper limb stroke rehabilitation. 3rd Int Conf on rehabilitation engineering & assistive technology; Singapore. ACM 2009; pp 1-4.
33. Zhang J, Ong SK, Nee AYC. Design and development of a navigation assistance system for visually impaired individuals. 3rd Int Conf on rehabilitation engineering & assistive technology; Singapore. ACM 2009; pp 1-4.
34. Raichur A, Wihardjo G, Banerji S, Heng J, editors. A step towards home-based robotic rehabilitation: An interface circuit for EEG/SEMG actuated orthosis. IEEE Int Conf on advanced intelligent mechatronics 2009; pp1998-2003.
35. Mazzoleni S, Toth A, Munih M, et al. Whole-body isometric force/torque measurements for functional assessment in neuro-rehabilitation: platform design, development and verification. J Neuroeng Rehabil. 2009; 6: 38.
36. Sasidhar S, Panda SK, Xu J. A real time control algorithm for a myoelectric glove for the rehabilitation of wrist and elbow of stroke patients. 8th IEEE Int Conf on control and automation 2010; pp 745-749.
37. Kikuchi T, Ozawa T, Fukushima K, et al. Development of Force-Measurable Grip and Software for "PLEMO", Rehabilitation System for Upper Limbs Based on Physical Therapy. Int Conf on service robotics and mechatronics 2010; pp 81-86.
38. Goto S, Nakamura M, Sugi T. Development of meal assistance orthosis for disabled persons using EOG signal and dish image International Journal of Advanced Mechatronic Systems 2008; 1(2): 107-115.
39. Al-Jumaily A, Olivares RA. Electromyogram (EMG) driven system based virtual reality for prosthetic and rehabilitation devices. 11^th Int Conf on Information integration and web-based applications & services; Malaysia. ACM 2009; pp 582-586.
40. Nagaoka T, Sakatani K, Awano T, et al. Development of a new rehabilitation system based on a braincomputer interface using near-infrared spectroscopy. Adv Exp Med Biol 2010; 662: 497-503.
41. Li H, Song A, Zhang H. Development of a forceassistant tele-rehabilitation system for the stroke. IEEE Int Symposium on industrial electronics 2007; pp 1360- 1364.
42. Nef T, Mihelj M, Riener R. ARMin: a robot for patientcooperative arm therapy. Med Biol Eng Comput 2007; 45(9): 887-900.
43. Ahamed NU, Sundaraj K, Poo TS. Design and development of an automated, portable and handheld tablet personal computer-based data acquisition system for monitoring electromyography signals during rehabilitation. Proc Inst Mech Eng Part H-J Eng Med 2013, 262-274.
44. Costin H, Cehan V, Rotariu C, Morancea O, Felea V, Alexa I, et al. TELEMON - A Complex System for Real Time Telemonitoring of Chronic Patients and Elderly People. 4th European Conf of the Int Federation for Medical and Biological Engineering: Springer Berlin Heidelberg 2009; pp 1002-1005.
45. Dobrescu R, Dobrescu M, Popescu D, Coanda HG, editors. Embedded Wireless Homecare Monitoring System. eHealth, Int Conf on Telemedicine, and Social Medicine 1-7 Feb 2009; pp 66-71.
46. Son JS, Kim JY, Hwang SJ, Kim Y. The Development of an EMG-based Upper Extremity Rehabilitation Training System for Hemiplegic Patients. 13th Int Conf on Biomedical Engineering: Springer Berlin Heidelberg 2009; pp 1977-1979.
47. Beyl P, Cherelle P, Knaepen K, et al. A proof-ofconcept exoskeleton for robot-assisted rehabilitation of gait. European Conf of the Int Federation for medical and biological engineering 2009; pp1825-1829. 48. EEG Signal Processing. http://www. controlsystemslabgr/ index/neuroeqhtml.
48. Ino S, Sato M, Hosono M, Izumi T. Development of a soft metal hydride actuator using a laminate bellows for rehabilitation systems. Sens Actuator B-Chem 2009; 136(1): 86-91.