ROBOTIC REHABILITATION
Robotics technology has evolved remarkably in recent years, with faster and more powerful computers and new computational approaches, as well as increased complexity of electro-mechanical components. This advancement in technology has made robots available for rehabilitation interventions. A robot is defined as a reprogrammable, multifunctional manipulator designed to move material, parts, or special devices with variable programmed movements to perform a task (Chang & Kim, 2013) .
Robot-mediated rehabilitation is an innovative, exercise-based therapy using robotic devices that allows the implementation of highly repetitive, intense, adaptable and measurable physical training. Since the first clinical studies with the MIT- Manus robot, robotic applications, especially in the upper extremity stroke patients with paresis , spinal cord injury, multiple sclerosis, as well as cerebral It is increasingly used to correct loss of motor function in palsy . Therefore, many studies have suggested that robot-assisted training integrated into a multidisciplinary program provides an additional reduction in motor impairments compared to usual care alone at different stages of stroke recovery (acute, subacute chronic) (Duret et al., 2019) .
Robotic devices can be used to facilitate passive normal joint movement, help maintain motion and flexibility, temporarily reduce hypertonia , and provide resistance during passive movement. Assistance can also be provided during active movements when the patient is unable to complete a movement independently. Robotics may be most suitable for patients with intense hemiplegia , but robotics can also be used in senior patients who want to increase strength by providing resistance during movement (Pauli et al., 2019) .
Operating the robot refers to the way energy is generated and involves the use of an electric motor, hydraulics, pneumatics or human muscle. Transmission refers to the way the robot transfers the movement of the actuator to the movement of the arm and includes connections and cables ( Yue et al. 2017 ). Finally, sensors detect the force and position of the upper limb to provide feedback in response , including physical or bioelectric signals such as electromyogram (Pauli et al., 2019) .
The field of rehabilitation robotics is generally divided into two general topics: robotics for aid and assistance (assistive robotics) and robotics for therapy and recovery ( therapeutic robotics) (Sadeghnejad et al., 2023) .
Assistive robots are systems and devices that help patients with mobility impairments due to age, injury or stroke move their bodies and perform daily tasks easier and faster. However, these robots do not necessarily engage the patient's nervous system to regain control of the disabled or paretic limb or provide any feedback about the patient's recovery. Since the application of such robots to many patients is long-term, even permanent, these robots must be carefully designed to be very useful. A well-known example of this group is the wheelchair ( Sadeghnejad et al., 2023 ) .
In contrast, therapeutic systems play at least one role in facilitating patients' recovery. These systems are often used to focus on improving a specific function of the patient for a predetermined and limited period of time. Therapeutic robots can work alongside a therapist or patient's caregivers, resulting in faster therapy for the patient and less fatigue and tension for the third party. Therapeutic robotic systems used in the field of neurorehabilitation can be grouped under two basic categories in the upper and lower extremities : exoskelaton and end - effector type robots ( Sadeghnejad et al., 2023) .
Therapeutic Robot Systems
History of Robotic Rehabilitation
Computer systems and robots have come in handy to assist humans in many areas over the last 50 years. Robots have assisted or completely replaced humans in a wide variety of jobs, some as easy as everyday human tasks, some as heavy as advanced healthcare systems, sports, hazardous industrial and agricultural processes, and life-threatening situations such as firefighting or military operations. A crucial invention that is driving the advancement of robotics is artificial intelligence (AI). As artificial intelligence evolved, automation technology and machine learning approaches were added to machines. This has led to smart equipment and efficient production with minimal human intervention (Sadeghnejad et al., 2023) .
The idea of using machines for rehabilitation dates back much further. Theodor in a 1910 patent Büdingen proposed a 'movement improvement apparatus', a machine driven by an electric motor, to guide and support stepping movements in patients with heart disease (Gassert & Dietz, 2018) .
In the 1930s, Richard Scherb developed the 'meridian', a cable-powered apparatus that moves joints for orthopedic treatment. This human-powered mechanotherapy machine supported multiple modes of interaction, from passive to active-assisted and active-resistance movements (Gassert & Dietz, 2018) .
The first robotic rehabilitation system was based on the concept of continuous passive motion (CPM), a rigid mode of interaction in which the robot moves joints along a predefined trajectory independent of the patient's contribution (Gassert & Dietz, 2018) .
The first powered exoskeletons for therapeutic applications in neurological patients were introduced in the 1970s. (Gassert & Dietz, 2018) . A new era in neurorehabilitation robotics began in 1989 with the development of MIT- MANUS , which was first tested clinically in 1994 . Compared to industrial manipulators, this planar manipulandum offers inherently low mechanical output impedance (a frequency-dependent resistance). ( motion detected at the interface between the human user and the robotic system ) and provides the upper limb against gravity, thus allowing the support to be adapted according to the severity of the deficits. (Gassert & Dietz, 2018) .
Effector- based devices for planar (MIT-MANUS) and 3D ( Gentle /S) reaching movements were introduced to allow more active patient participation . Later developments focused on incorporating additional degrees of freedom regarding wrist and hand open/close function ( Gentle /G). ARMin, an electric exoskeleton that also integrates grasping and releasing function, was developed for functional training of three-dimensional arm movements with guidance in three proximal joints . Regardless of their kinematic configuration , all of these systems are to partially or completely eliminate the arm against gravity. This approach, flexor It reduces the effect of synergies and allows hand movements to be performed in a larger work area. (Gassert & Dietz, 2018) .
Rehabilitation robots for the lower extremity began in 1994 with the design of Lokomat , which combined body weight supported treadmill training (BWSTT) with the help of a robotic gait orthosis . Gait Trainer implemented a similar concept based on end effector design (Gassert & Dietz, 2018) .
upper limb rehabilitation robots
lower limb rehabilitation robots
Upper Extremity Robots
Interaction with the environment occurs mainly through the hands and produces somatosensory feedback. However, after central nervous system damage, somatosensory function is often impaired. Therefore, neurorehabilitation devices for the upper extremity should train hand and, as far as possible, finger function by providing both visual and tactile feedback. Training should include tasks functionally related to ADL, such as grasping and releasing objects with virtual dynamics, to also train somatosensory function and sensorimotor integration (Gassert & Dietz, 2018) .
Exoskeleton ( Exoskelaton ):
Interest in this type of robots has increased in recent years. Exoskeleton robots are devices that provide support to the musculoskeletal system and movement when worn by patients, apply mechanical force by completely wrapping the extremity , and can move synchronously with the patient . Thanks to the synchronization between the patient and the robot, the robots detect the patient's movement ability , follow it, support its movement in line with the need, and increase its physical capacity and working time. In accordance with its intended use, it has purposes such as increasing strength and assisting movement (Molteni et al., 2018) .
They are similar to human limbs in that they are attached to patients at multiple points and their joint axes match human joint axes. It is possible to train specific muscles by controlling human movements at calculated torques . In short; It allows us to accurately determine the kinematic configuration of human joints . It can be fixed or ambulatory . Thus, the emergence of abnormal postures and movements is minimized (Molteni et al., 2018) . The main ones are described below.
ARMin
The robot can be used for right and left arm. A graphic display presents the patient with different training scenarios. These are passive mobilization , active play therapy and active daily living activities training. The device detects how much the patient contributes to the movement and provides support as needed (Pauli et al., 2019) .
Armeo Power :
3- dimensional workspace, allowing patients to perform intense, repetitive and targeted exercises. It is done through various motivating exercises, games and simulations. The device offers the patient the option of assisting or active participation, depending on the patient's needs. It is one of the most advanced devices with 6 degrees of freedom (Pauli et al., 2019) .
End Effector :
patient limb While we can control the proximal part , the functions in the distal part are supported by devices. The main advantage of such robots is that they can easily adapt to patients with different body sizes (Molteni et al., 2018) .
to the patient from a distal point, and the robots' joint areas do not match the patients' joints. The force generated at the distal interface changes the positions of other joints simultaneously, making isolated movement of a single joint difficult. limb When proximal control is inadequate, abnormal postures and movements occur (Molteni et al., 2018) . The main ones are described below.
MIT Manus :
MIT- Manus is one of the first robotic devices developed and is the most widely used end -effector system robot that works integrated with virtual reality technology and aims to perform exercises between 400 and 1000 repetitions . It is very easy to use and wear. It is a 2-DOF robot manipulator with visual, auditory, and tactile feedback that provides exercises to the upper extremity while the patient plays a series of video games that involve positioning the robot end effector. It allows flexion and extension movements in the shoulder and elbow . In addition, it also allows reaching movements in the horizontal plane (Pauli et al., 2019) .
MIME ( Mirror images motion Enabler )
single-sided modes, a two-sided mode similar to a mirror image has been developed. It is produced using robotic systems to assist the affected arm by mirroring the movements of the unaffected arm . It is the first robot to perform activities and has 6 degrees of freedom (Pauli et al., 2019) .
Gentle / S Heptic Master
It has 3 degrees of freedom. The patient is seated in a chair with his arm positioned on an elbow orthosis suspended from the upper circumference . This is done to eliminate the influence of gravity and prevent shoulder subluxation . wrist, haptic It is placed on a wrist orthosis connected to the interface . The physiotherapist selects the exercise and it is performed together with the patient ( Pauli et al., 2019 ) .
arms Guide
It is a single-controlled, four-degree-of-freedom robotic device. in the upper extremity It evaluates tone , spasticity and coordination (Pauli et al., 2019) .
Amadeo Robot
-effector- based system that has five degrees of freedom and enables movement of one or all of the five fingers through a passive rotation joint placed between the fingertip and a laterally moving device (the thumb has two passive rotation joints) . All five degrees of freedom are independent and cover almost the entire working area of the fingers. The interface between the human hand and the machine is provided by elastic bands or casts, and the movement of the wrist is prevented by a velcro strap (Sale et al., 2014) .
Lower Extremity
It combines mechanical power devices and artificial intelligence to assist patients with lower extremity dysfunction and improve muscle mobility leading to repair of physical impairments (Sadeghnejad et al., 2023) . During rehabilitation, physical support needs to be constantly adapted to the patient's actual needs by reducing and selectively providing assistance in order to maximize the patient's active participation (Gassert & Dietz, 2018) .
Exoskeleton ( Exoskelaton ):
the extremity externally and are used for diagnosis and treatment purposes . The suspension system supports the body weight and provides the patient with correct functional mobility . provides the opportunity (Pauli et al., 2019) .
lokomat
a body weight support system, robotic walking orthosis and treadmill . There are drivers that provide synchronization between the walking orthosis and the treadmill speed . Walking orthosis works by computer-controlled motors that direct the movement of the hip and knee during walking. It is one of the most used rehabilitation robots to reinforce motor and cognitive abilities in walking patterns (Pauli et al., 2019) .
AutoAmbulator
a body weight support system, robotic walking orthosis and treadmill . Robotic arms are fixed at the thigh and ankle. It offers personalized exercises by arranging a natural walking pattern to ensure balance, coordination and endurance (Pauli et al., 2019) .
ALEX
Its working principle is that unwanted walking movement is prevented and assistance is provided for the desired movement (Pauli et al., 2019) .
KineAssist
The system, which has a mobile robotic base, was designed considering the risk of individuals falling. It allows trunk movements with trunk and pelvis mechanisms, and while these mechanisms provide a natural walking pattern , they enable the patient to feel safe without falling (Pauli et al., 2019) .
walk Trainer
It is a mobile device that provides training in upright posture and physiological movement for stroke patients by closely monitoring natural, voluntary, over-ground walking. Due to its limited range of motion , the hip does not allow abduction . For this reason, patients have some difficulty in terms of balance (Pauli et al., 2019) .
Hybrid assistive Limb –Hal ( Cyberdyne I Unc , Japan)
It has been developed for support in daily living activities and heavy work, apart from rehabilitation applications . (Pauli et al., 2019) .
End effector ( end - effector ):
of robots proximal While they can control the parts themselves, the devices used support the functions in the distal part. In lower extremity robots, the patient's feet are placed in foot pedals and different walking patterns are revealed (Pauli et al., 2019) .
Gangtrainer (GT)
By carrying the patient's body weight by the device and adapting the movement speed to the patient's individual ability, it helps the patient regain his mobility . Patients are placed on two foot plates that stimulate the gait and swing phases , and ropes attached to the patient can control vertical and lateral movements of the center of mass. It is considered one of the pioneering robotic systems for rehabilitation (Pauli et al., 2019) .
haptics walker
which includes partial body weight support and programmable foot pedals, stimulates walking not only on slow and smooth surfaces but also on rough and slippery surfaces. It does. It is designed at a more detailed level than gangtrainer (Pauli et al., 2019) .
GaitMaster 5
Patient feet are placed on pads covered with sensors . connects.this The pads are connected to movement platforms that can move the foot forward or up and down (Pauli et al., 2019) .
LokoHelp
Lokohelp is placed and fixed in the middle of the walking belt surface, parallel to the walking direction . It also includes a body weight support system and foot pads . An improved walking orthosis It requires less therapeutic assistance and pelvic movements are provided by the patient (Pauli et al., 2019) .
RESOURCES
Chang, W.H., & Kim, Y.-H. (2013). Robot-assisted Therapy in Stroke Rehabilitation. Journal of Stroke , 15 (3), 174–181. https://doi.org/10.5853/jos.2013.15.3.174
Gassert, R., & Dietz, V. (2018). Rehabilitation robots for the treatment of sensorimotor deficits: A neurophysiological perspective. Journal of NeuroEngineering and Rehabilitation , 15 (1), 1–15. https://doi.org/10.1186/s12984-018-0383-x
Pauli, G., Iruthayarajah, J., Mirkowski, M., Ot, M., Ont, R., Iliescu, A., Caughlin, S., Fragis, N., Harris, J., Dukelow, S., Chae, J., Knutson, J., Miller, T., & Teasell, R. (2019). Chapter 10: UPPER EXTREMITY ENGINE . 1–366.
Sadeghnejad, S., Abadi, VSE, & Jafari, B. (2023). Rehabilitation robotics: History, applications, and recent advances. Medical and Healthcare Robotics , December 2022 , 63–85. https://doi.org/10.1016/b978-0-443-18460-4.00008-1
