Application of Mobile-Based Games in The Rehabilitation of Stroke Survivors

Background

A stroke happens due to an interruption in the brain’s blood supply (Baranyi, Czech, Hofstätter, Aigner, & Grechenig, 2020). Impairment in oxygen delivery causes brain cells to be injured (Baranyi et al., 2020; E. Wilkins et al., 2017). Stroke is classified as ischemic or hemorrhagic, which is due to either blockage or rupture of an artery in the brain, respectively. Transient ischemic attacks or mini-strokes also occur in which blood flow is hampered for a short period (Baranyi et al., 2020). The incidence of stroke has been increased globally, particularly in the middle- and low-income countries, because of the growing prevalence of some risk factors like aging, smoking, hypertension, alcohol consumption, insufficient physical activity, raised lipid profile (Baranyi et al., 2020; Katan & Luft, 2018; E. Wilkins et al., 2017). Elevated incidence of stroke shows the alarming trend of people with functional impairments (R. W. Teasell, Foley, Salter, & Jutai, 2008).

Different types of deficits like aphasia, weakness, sensory loss, and cognitive disorders emerge in stroke survivors. Cognitive and upper-limb impairments are more prevalent than communication and post-stroke depression (Nakayama, Jørgensen, Raaschou, & Olsen, 1994; Salter et al., 2005; R. Teasell, Salter, Faltynek, Cotoi, & Eskes, 2018). In most cases, survivors of a stroke suffered permanent disability. They are subjected to rehabilitation therapy to reestablish the brain’s ability via rerouting the neural pathways surrounding the damaged area (Nudo, 2007) to restore the lost skills or development of compensating ways to accomplish the required task (Carabeo, Dalida, Padilla, & Rodrigo, 2014).

Motor disorders are one of the major problems in stroke patients. They are defined as a loss or limited muscle control or movement. Nearly 80% of patients lose control of the face, arm, and leg movement on one side. The treatment plan for such patients is mainly focused on rehabilitation exercises intended to improve impaired movement and relevant functions (Langhorne, Coupar, & Pollock, 2009). However, only one-fourth of the patients reach full recovery by physical functioning programs, (Dobkin, 2005) possibly because conventional rehabilitation programs are not easy-to-use processes. Stroke patients with spasticity or upper and lower extremity dysfunctions undergo rehabilitation as well. Recovery is often seen three months after stroke in these patients. It takes, however, several years in some cases (Hebert et al., 2016).

Many stroke patients experience upper-extremity impairments for long-term periods, which impose severe restrictions in daily activities (Faria-Fortini, Michaelsen, Cassiano, & Teixeira-Salmela, 2011; Kizony, Zeilig, Dudkiewicz, Schejter-Margalit, & Rand, 2016). The primary motor cortex of the brain controls fine motor skills, which means the coordination of small-muscle movements toward acting delicate, complex, and skilled performances by our fingers. Loss of control in this region makes simple tasks a labor one (Carabeo et al., 2014). They also need rehabilitation exercises for hands and fingers to regain normal functions. In subacute or chronic phases, improvement of upper-extremity limbs is hampered if patients do not adhere sufficiently to the training intensity and treatment plans (Kaur, English, & Hillier, 2012; Lang, MacDonald, & Gnip, 2007; Lang et al., 2009).

Rehabilitation has become an impactful aspect of health improvement in stroke survivors (Saposnik et al., 2014). Although physiotherapy and occupational therapy, as two types of conventional rehabilitation approaches, improve post-stroke motor function, there are also some limitations (Jutai & Teasell, 2003; Langhorne et al., 2009; R. W. Teasell et al., 2008). These include a shortage of provider centers and human resources, leading to insufficient rehabilitation in terms of intensity and duration (Jutai & Teasell, 2003; R. Teasell et al., 2009; R. W. Teasell et al., 2008). Routine rehabilitation programs are time-consuming, labor and resource-intensive, and the outcomes depend on patient adherence. Geographic location also determines availability. Specifically, stroke survivors do not initially show the improvements, and therapeutic benefits are modest or even appear with a delay. For instance, repetitive and intense training, which are requirements of fine motor recovery, are not provided by a one-hour outpatient program. Considering the fact that discharged patients have limited rehabilitative alternatives for home use (Saposnik et al., 2014), there is an intense need for innovative strategies, especially those of self-training, that promote patients' motivation (Kizony et al., 2016).

Promising advancements in software technology and neurobiological investigations are incredibly improving strategies of stroke rehabilitation. These may be invasive approaches like using stem cells/neural precursors, and creating a brain-machine interface via implanting electrodes to drive or read a circuit. Of non-invasive approaches are recent innovations in virtual reality, robotic tools, stimulation by transcranial direct current, and stimulation through repetitive transcranial magnet (Owolabi et al., 2020). For instance, type, quantity, and quality of practices could be monitored remotely via tele-rehabilitation in the home or community setting. Also, therapist can make necessary feedback in order to optimize the frequency, intensity, and progression of the trainings (Cramer, Dodakian, & Le, 2019). This method is an outstanding one especially for underserved regions to gain access to high quality care. In the present study, some of the latest innovations in the era of mobile applications that are developed in order to use in stroke rehabilitation were reviewed.

Methods

Databases including PubMed, Scopus, and Embase were searched against keywords. Rehabilitation, Stroke, Mobile/Computer, and Application/Game were used as keywords separately or in combination with each other. Inclusion criteria were studies in which stroke survivors (irrespective of its type: general, hemorrhagic, ischemic, and chronic) use novel mobile applications as a part of their rehabilitation program. Those papers in non-English language were excluded. All retrieved articles were imported to an Endnote file, and after deleting duplicates, their titles were screened. Discarding irrelevant articles, abstracts of the remaining were carefully read, and similarly, non-useful ones were omitted. Finalized articles were completely reviewed and relevant information, which are in line with the aim of our study, were extracted.

Results

- Use of modern technology for rehabilitation of stroke patients

Conventional and modern approaches of stroke rehabilitation are different in many aspects. Mainly, conventional exercises are inferior with respect to using modern technology, and traditionally classified into neurodevelopmental therapy and task-oriented approaches (Nair & Taly, 2002). While task-oriented practices such as task-specific arm rehabilitation, constraint induced movement therapy, and body weight-supported treadmill training are only feasible in the presence of motor control, current high-tech strategies are designed in an impairment-oriented manner, which are indeed practical in real-world condition in the case of disorder in motor control. These include, but not limited to, arm basis training, mirror therapy, arm rehabilitation via robots, neuromuscular electrical stimulation, etc (Owolabi et al., 2020).

Modern technology has opened new avenues in rehabilitation of stroke patients. It is reflected in introducing robotic devices like exoskeleton, virtual reality, augmented reality like interaction techniques, and serious games. Training tasks of these products could be individualized based on the user own interests leading to a long-lasting motivation, positive outcomes, and swift recovery (Tageldeen, Elamvazuthi, Perumal, & Ganesan, 2017). Although there is a consensus on rehabilitation benefits for walking function by robotic devices in stroke survivors, such approaches need to be optimized in order to improve neurorecovery.

Exoskeletons are external devices that are designed to give support to a part of human skeleton. With links similar to joints, torques are transmitted from exoskeleton structure to the body by the aid of actuators (Perry, Rosen, & Burns, 2007). Effectiveness of this type of rehabilitation tool was corroborated in several studies (Burke et al., 2010; Grimm, Naros, & Gharabaghi, 2016; Rahman et al., 2015; Schabowsky et al., 2010). In a pilot interventional study, a grounded exoskeleton combined with a video game visual feedback was used to promote ankle movements. Task difficulty was automatically modulated according to the received performance. It was demonstrated that task performance and motor outcomes became better in the target muscles of both stroke patients and healthy peers. Although excitability was not similar in different parts of the corticospinal tract, functional improvements were reported in total. The study showed that this robot-assisted approach improved motor learning as well. Such robotic training tools are potentially interesting for gait rehabilitation in post-stroke settings (Asín-Prieto et al., 2020).

As attenuated sensation, spasms, loss of muscle tone, and curling of the fingers are among the important symptoms of stroke-induced paralysis (Gray, Rice, & Garland, 2012), rehabilitation of the hands is a complex task. Fortunately, plasticity of the human neurons makes learning and memory possible (Yue, Zhang, & Wang, 2017). In other words, it is feasible to remap the functions of motor cortex through repetitive motor exercises. In one study, a lightweight, compact, and affordable exoskeleton that is designed on a drive-by-string mechanism and controllable by a mobile application was developed to assist flexion and extension motions of the fingers in order to restore voluntary functions. Findings revealed that required torque to flex each finger was reached with accuracy of 90% (Snekhalatha, Muppidi, & Akkiraju, 2022) .

The area of robotics (like Armeospring, and MIT Manus) is the latest progress in the area of post-stroke rehabilitation with promising advantages like easy-to-use, affordability, automation, accurate control in a wide range of motions, and enhanced degrees of freedom compared with conventional methods (Daly et al., 2005). However, this type of technology also has some disadvantages such as being hefty with complicated systems.

Virtual reality is in fact the enhanced version of interaction between human and computer in a non-real environment. Earphones, Head-Mounted Displays, and last generations of gaming systems (Xbox 360 Kinect and Nintendo Wii) are some examples of such technology. It was shown that street program of virtual reality was effective to treat post-stroke patients (Yates, Kelemen, & Sik Lanyi, 2016). Stroke patients are encouraged to use virtual reality in either clinical settings or their own homes (Burdea, 2003). Virtual reality with superior sense of realism is called augmented reality. Reports demonstrated that hand stiffness was improved upon using of a haptic device with augmented reality (Khademi, Hondori, Lopes, Dodakian, & Cramer, 2012). The advantage of augmented reality is using real-world objects in a virtual environment (Burke et al., 2010).

Another example of using modern technology in stroke rehabilitation is gamification, which means using principles of game designs in non-game contexts. Games possess novel interaction modalities that support a promising and appealing experience in stroke survivors (Langan, Subryan, Nwogu, Cavuoto, & Technology, 2018). Nowadays, games/video games such as 2D (Loureiro et al., 2011), 3D (Morrow, Docan, Burdea, & Merians, 2006), virtual reality games (Lewis, Woods, Rosie, Mcpherson, & Technology, 2011), augmented reality games (Burke et al., 2010) as well as natural user interfaces like Wii (Saposnik et al., 2010), PlayStation, Wii Balance, Xbox, and Kinect (Roy, Soni, & Dubey, 2013) are routinely used for rehabilitation. As an important part of treatment plan, they play prominent roles in encouraging patients to be adhere to rehabilitation strategies (Mubin et al., 2022).

Serious games for rehabilitation provide motivation, encouragement, and individualized use of repetitive exercises along with receiving feedback that is augmented by boredom alleviation (Mubin et al., 2022). Motion-based games are favored as they provoke stroke patients to perform intended tasks in a routine manner without getting bored. In one study, design principles and characteristics of such games were reviewed. It was concluded that gaming is a very useful interactive approach for rehabilitation of stroke patients. Furthermore, these games could be customized and contextualized for specific situations. Also, possible challenges and errors are resolved by incorporating effective features into serious games (Mubin et al., 2022).

According to the Global Observatory for eHealth, any medical and public health practices that are executable on mobile devices such as cell phones, devices designed for patients monitoring, digital assistants of patients and other wireless tools are defined as mobile health (technologies., 2011). Such features are used in a variety of forms including video games (Zadro et al., 2019), gait training via music software (C.-H. Wang et al., 2014), and mobile applications (Geraedts et al., 2017). In this regard, mobile-based games are practical solutions that could be installed on mobile devices like mobile phones and tablets. Although feasibility and benefits of mobile health were confirmed through significant improvements in different disorders like diabetes and cardiac rehabilitation programs (Beatty, Fukuoka, & Whooley, 2013; Quinn et al., 2008), technical issues should be considered in each mobile device.

In this section, several commercial mobile applications specifically designed for the rehabilitation of stroke patients will be discussed. It should be noted that each application is used for a specific deficit. FINDEX is an android-based tablet game to enhance fine motor skills, excellent finger dexterity (Carabeo et al., 2014). The design of FINDEX is in accordance with daily household and work-related activities. It provides a dragging task for finger control, a tapping task for addressing finger isolation and coordination, and a stretching task for increasing the range of motion in the fingers (Carabeo et al., 2014).

REHA@STROKE is another mobile-based application that gives some rehabilitation exercises for stroke patients. This serious game could evaluate movement, touch, gesture, and cognitive categories via a mobile phone. For instance, pinch-like activities are done in the variety of motion for rehabilitation of fingers or touch category provides different tasks with fingers by tapping on the screen (Baranyi et al., 2020). There are three different levels in each category (except cognitive) that should be solved within a short period. For patients who complete all three levels in each category, an extra challenging step needs to be played in a much longer time. Developers thought that this approach brings great beneficial potential for stroke patients. However, the results should be further evaluated in a more significant number of patients during a more extended period. A control group needs to be considered in the study to compare the findings with the treated patients (Baranyi et al., 2020).

RECOVERNOW is a tablet-based platform designed to meet the specific needs of speech-language therapy and, or occupational therapy in stroke patients during the entire treatment course, from the acute care section up to the post-discharge period at home (Pugliese et al., 2019). In one study, 30 patients with acute stroke were recruited using RECOVERNOW. Each patient was asked to play at least one hour per day with the application. After three months, they were invited to participate in a follow-up interview. The follow-up interview was not for assessing the efficacy or checking improvements, but it was estimated the percentage of successful interviews for upcoming clinical trials. Feasibility as the primary outcome was evaluated based on five criteria, including recruitment rate, adherence rate, retention rate, the proportion of successful follow-up interventions, and protocol deviations. Indeed, patients’ depression was also screened through a questionnaire as severe depression affects the rehabilitation process adversely, and the depression should be treated first before rehabilitation programs are started. The participation rate in the follow-up interview was reported to be 77%. The authors reported easy recruitment and patients’ interest in using this platform. However, there were some patient-, device- and system-related hurdles. Depression was approved in 37% of the patients. Overall, the authors concluded that the patients are interested in starting tablet-based rehabilitation within the first week after stroke. However, the application should be optimized according to each patient’s specific deficit to minimize the hurdles and maximize the therapeutic efficiency (Pugliese et al., 2019).

An iOS-based mobile application, ARMSTROKES, supported elbow and shoulder exercises in stroke patients. The movement of the patients is recorded via integrated cell phone sensors (accelerometer, gyroscope, and orientation sensor). This very customizable application showed improvements in the patients (Lawson, Guo, Smith, Tang, & Feng, 2016). It was demonstrated that the portability of this type of rehabilitation allows persistent adherence to performing exercises in daily routines. It also elevates compliance and optimizes motor function, leading to more participation in a daily schedule. Other than reduced fatigue, the number and accuracy of movements, which indicates motor learning, were improved. Active and passive range of motions were also increased. Participants declared that their ability to carry out daily activities had improved. They were motivated to use the application as a part of home rehabilitation programs. The application improved occupational performance as well. This aspect should be scrutinized further to differentiate from naturally occurring improvement. However, these changes were not statistically significant, possibly due to the sampling method and goals of the study. Indeed, the authors considered the improvements might result from home exercise programs other than the application (Lawson, Tang, & Feng, 2017).

The iPad, with its elaborated sensitivity to the multi-touch interface, helps health experts to monitor patients' activities more accurately. One of the therapeutic applications in the iPad is DEXTERIA which improves fine motor skills. In particular, DEXTERIA uses many hand exercises, including tapping, pinching, and writing, to improve strength, control, and skill via promoting finger flexion, extension, talent, and isolation (Carabeo et al., 2014). DEXTERIA is a mobile application, which is increased fine motor and handwriting skills. It is usable for both children and adults (Carabeo et al., 2014). DEXTERIA was compared between stroke subjects with and without disability (Y.-C. Wang, Bohannon, Kapellusch, Garg, & Gershon, 2015). The overall experiment showed that tablet applications had positively affected hand improvements in stroke patients. The application was enjoyable for the participants, and they found it an effective tool in post-stroke rehabilitation (Kizony et al., 2016).

A randomized clinical trial reported the effects of iPad technology on the rehabilitation of patients after an acute ischemic or hemorrhagic stroke. Patients were asked to play the STROKE REHAB game on their own for 20 minutes per day in 10 days to 14 days. This game with six stages was specifically developed to improve acceptable motor weakness and, or neglect. Feasibility, as the primary outcome, was defined as the total playing time (max. 200 minutes). The game was feasible if patients played at least ≥140 minutes (Saposnik et al., 2014).

- Challenges of using mobile applications by stroke patients

Post-stroke injuries are different, and each specific game addresses only certain dysfunctions. Specifically, there is a lack of games to measure skill in fine motor functions, to quantify patients’ problems, and to determine the efficiency of such games (Carabeo et al., 2014). Some of these games are not appropriately designed for the rehabilitation of stroke patients. They lack some options in terms of adaptation to patients’ disabilities, and tracking and monitoring of motion (Borghese, Pirovano, Lanzi, Wüest, & de Bruin, 2013). Patients complain about some hurdles like the difficulty of the games (Pugliese et al., 2019). Some games are not easy-to-use for stroke patients because speed, coordination, and isolated finger movements are essential elements to complete the game’s steps, which are absent in stroke patients. Accordingly, transfer between different games and applications is inevitable based on the patients’ preferences (Kizony et al., 2016).

Using mobile-based games faces with some severe challenges. According to the nature of these challenges, they could be categorized into device-, patient-, and system-related (Pugliese et al., 2019). Of the first one are functionality of the applications in different phone models, poor screen touch of the phones, unresponsiveness of the phones’ configuration, boring contents, in-applications’ advertisements, language barrier, and programming errors. Patient-related issues are eyes’ tiredness, difficulties in reading, focusing, and instructions besides fine-motor hurdles, hand numbness and jitter, training steps, and interruption in use. Making a suitable communication is an example of the system-related challenges.

Applications were different in a variety of aspects. While some of them were designed for assessing (stroke, severity, language, motor deficits, etc.), others were developed for training (lower limb and balance, upper limb improvement, and for certain skills such as visual, cognitive, and speech). There are some other applications focused on compensatory strategies (Marwaa, Guidetti, Ytterberg, & Kristensen, 2022). Also, some applications are designed for supporting health-related information or education and discharge (Epalte, Tomsone, Vētra, Bērziņa, & Technology, 2020; Kang et al., 2019; Markle-Reid et al., 2020; Siegel et al., 2016; Sureshkumar, Murthy, Munuswamy, Goenka, & Kuper, 2015), and other ones support participation, setting goals, and continuous evaluation (Markle-Reid et al., 2020). In the case of comparison between applications, it should be kept in mind that associated studies are limited, have small sample size, and conduct within dissimilar design type. Other than comparison, these conditions limit their clinical efficacy (Marwaa et al., 2022). Other factors like chronic vs. acute/subacute stroke, and rehabilitation setting (at homes or clinical centers) may further affect the generalizability of the findings. For example, majority of the applications provide only one aspect of support (assessment or training) (Fell et al., 2019) while there are some others that simultaneously manage deficits, discharge needs, and support exercise programs (Sureshkumar et al., 2015).

Discussion

- Stroke survivors and rehabilitation

Cognition, visual, or motor abilities are significantly impaired in most of the stroke survivors. This imposes a huge burden on healthcare services, which in turn deteriorates providing required advocacy. Only 15% of stroke patients reach complete motor recovery in upper and lower extremities (Hendricks et al., 2002). Conventional rehabilitation do not give sufficient number of movements that are required to reach the satisfactory goal. Precise and repeatable therapeutic exercises could be performed by the use of rehabilitation devices that have also great benefits such as physiotherapist-independence, collecting quantitative data, and supporting feedback mechanisms accordingly (Lee et al., 2018).

- Rehabilitation via mobile games

Most studies, which have been discussed previously, have focused on the rehabilitation of upper limbs. Vendors tried to present target-oriented games with interesting content in which stroke survivors enthusiastically played by involving their fingers and hands and made efforts to reach the following levels. Also, some games have been specifically designed to replenish lost speech skills. Each level needs certain movements to be completed to overcome specific deficits. All these games could be installed on mobile devices like smartphones and tablets that are readily accessible by the patients.

- Rehabilitation at the neuronal level

Neuroplasticity is a paramount factor for motor recovery in stroke survivors. Neuroplasticity is the ability of the nervous system to adapt or regenerate after injury. Recovery of motor functions is commenced after the acute stage of stroke through neuroplasticity (Forrester, Wheaton, & Luft, 2008). Findings confirmed the capacity of neuroplasticity for rehabilitation of motor deficits after stroke (Arya, Pandian, Verma, & Garg, 2011). Cortical changes, which are happened after motor learning strategies, repetition, goal-directed and task-specific exercises, targeted challenges, and intensive works, are required for neuroplasticity to develop new motor pathways (Daly & Ruff, 2007). Neurological recovery and adaptation and the generation of alternative routes are essential determinants to recover motor skills. These facilitate motor relearning, followed by functional improvement, possibly due to cortical reorganization and brain plasticity (Barreca, Wolf, Fasoli, & Bohannon, 2003; Kwakkel, 2006; Luft et al., 2004).

The most well-known therapeutic phenomenon in post-stroke settings is known as neurorehabilitation. To reach this goal, high-intensity physiotherapy and constraint-induced movement therapy are the most beneficial interventions (Langhorne et al., 2009). Current guidelines recommend starting rehabilitation 5-7 days after stroke (Heart, 2014). However, the lack of experts and shortage of required facilities postpone this goal (Bernhardt, Dewey, Thrift, & Donnan, 2004; Control & Prevention, 2007; Heart, 2014; is Function, 2007). On the other hand, conventional rehabilitation approaches are tedious with low immediate improvements that require transportation to the rehabilitation centers, which is hard for stroke patients (Jutai & Teasell, 2003; Saposnik et al., 2014; R. Teasell et al., 2009; A. R. Wilkins, Kurland, & Stokes, 2013).

The extent of functional motor recovery after stroke is in association with the activity of neurons in the sensorimotor cortex of the ipsilesional hemisphere (Allman et al., 2016; Calautti & Baron, 2003; Mintzopoulos et al., 2008). Cortical reorganization and functional recovery is promoted in ipsilesional area upon repeating sensorimotor tasks through increasing bilateral cortical activity, which enhances neuroplasticity (Khanicheh et al., 2008).

- Rehabilitation interventions

The most usual rehabilitation interventions in stroke are strengthening and endurance induction of the muscles (Patten, Lexell, Brown, & Development, 2004), forced-use therapy (Liepert, Uhde, Gräf, Leidner, & Weiller, 2001), constraint-induced exercise (Schaechter et al., 2002), robotic tools equipped with biofeedback (Cordo et al., 2013), nonparetic limb interventions (Oujamaa et al., 2009; Sathian et al., 2000), and bilateral/bimanual approaches (Guerin et al., 2012; Wolf et al., 2014).

A motor recovery intervention combined with sensory feedback, motor control, and motor intention was assessed for rehabilitation of impaired unilateral arm after stroke. The authors reported that a particular self-guided motion (self-supported exercise) is able to reactivate pattern of healthy muscle in the paretic arm. In fact, nonparetic arm supports movement of the paretic one. In this unilateral stroke-induced motor paralysis, the efficiency of recovery time was improved via a biological mechanism. Speed, extent, and accuracy of stroke recovery significantly became better upon self-supported exercise. This shows that appropriate contralesional circuits is an important therapeutic point for damaged neural pathways. However, effective self-support therapy needs simultaneous activation of local neural circuits and top-down sensorimotor intention pathways. Other areas affected by stroke could be the future targets of self-support paradigm (Alnajjar et al., 2020).

- Novel approaches for rehabilitation:

In this regard, home-based rehabilitation emerged with advancements in technology and opened new avenues for goal-oriented and motivating training steps. Tablet computers with relevant applications rely on their inherent properties like portability and inexpensiveness (Kizony et al., 2016). The widespread use of smartphones and similar devices is an excellent opportunity to provide rehabilitation for different types of deficits in stroke patients like communication (Choi, Park, & Paik, 2016; Crotty et al., 2014; Des Roches, Balachandran, Ascenso, Tripodis, & Kiran, 2015; Hoover & Carney, 2014; Lavoie, Routhier, Légaré, & Macoir, 2016; K. H. Mallet et al., 2016; Routhier, Bier, & Macoir, 2016; Stark & Warburton, 2018; A. R. Wilkins et al., 2013), cognition (Des Roches et al., 2015; Katalinic, Young, & Doolan, 2013; K. Mallet et al., 2019), and fine-motor skills (Carabeo et al., 2014; Jang & Jang, 2016; Kizony et al., 2016; Rand, Schejter-Margalit, Dudkiewicz, Kizony, & Zeilig, 2013). One of the benefits of mobile-based therapies is the feasibility of its use in acute care section, where conventional facilities for rehabilitation are absent (Bernhardt et al., 2004). Patients with severe traumatic brain injury received benefits concerning attention training from interactive visuohaptic technology (Dvorkin et al., 2013). Another study claimed that such applications were helpful in improving communication and clinical goals (Tomori et al., 2012).

Novel technologies like robotics and virtual reality showed promising effects (Laver et al., 2017; Saposnik et al., 2014). Rehabilitation concepts like repetition, intensity, and task-oriented exercises are represented in new technologies like tablet-based games. Studies confirmed that virtual reality games improve motor function after stroke as high as approximately five times (Henderson, Korner-Bitensky, & Levin, 2007; Saposnik et al., 2014). Virtual reality accelerates cortical reorganization and brain plasticity (Jang et al., 2005). Imitation-related learning mirrors neural networks inducing organization around the motor cortex (Buccino, Solodkin, & Small, 2006). Also, findings reported that playing video games increases the volume of the ventral striatum i.e., mesolimbic pathways of the brain reward system (Kühn et al., 2011). In the same way, tablet technology uses multi-sensorial (visual, auditory, and tactile) feedback, affordability, and easy implementation. However, little data is available on these technologies in stroke patients (Saposnik et al., 2014).

Simulation/game designers work for the production of rehabilitation programs through creating health-related games (Lane, Slavin, & Ziv, 2001; Streufert, Satish, & Barach, 2001; Thompson et al., 2010). Tablet computers and game platforms were found helpful for rehabilitation in different aspects, including but not limited to entertainment, sense of success and achievement, and motivation for improvement (Whitcomb, 1990). The multi-touch display of these devices makes it possible to sense different movements like taps, swipes, pinches, or flicks transforming into real actions. Commercial virtual reality platforms like iPad, Nintendo Wii™, and Kinect™ are accessible and practical for use at the home (Saposnik et al., 2014). Also, patients declare numerous positive reports by using mobile-based therapies such as high satisfaction (Carabeo et al., 2014; Choi et al., 2016; Crotty et al., 2014; Kizony et al., 2016; K. Mallet et al., 2019; K. H. Mallet et al., 2016; Rand et al., 2013; White, Janssen, Jordan, & Pollack, 2015; A. R. Wilkins et al., 2013), therapy independence (Carabeo et al., 2014; White et al., 2015; A. R. Wilkins et al., 2013), and home-based rehabilitation (Crotty et al., 2014; A. R. Wilkins et al., 2013).

Game devices like Nintendo Wii are suitable tools because of the feasibility of performing goal-directed movements repetitively with appropriate challenges, and eventually receiving feedback on the performance (Arya et al., 2011; Saposnik et al., 2014). The interesting point is participants' excitement and application adaptability to provide the needs of stroke patients with different degrees of disease severity. The major benefits of motor training by this approach include promoting neuroplasticity and ease of access for underserved populations (Lawson et al., 2017). However, the high cost of these devices limits their use. Although home exercise programs are designed for stroke survivors as an extension of conventional interventions, certain restrictions like poor compliance, and self-efficacy, besides low motivation and perceived benefit, are reported (Chen, Neufeld, Feely, & Skinner, 1999).

In particular, serious games are a kind of non-entertainment activity with an enormous potential to motivate rehabilitation in stroke patients (Rocha, Rego, Faria, Reis, & Moreira, 2016). Serious games are the ones that are designed for a particular aim (Freitas & Liarokapis, 2011; Girard, Ecalle, & Magnan, 2013; Tsekleves, Cosmas, & Aggoun, 2016). Serious games are diverse interventions intended for motor recovery that offer entertainment and therapeutic benefits to stroke patients (McCallum, 2012). It is indeed more appealing and motivational when the rehabilitation process is in game based-programs in patients’ own homes (Hung, Huang, Chen, & Chu, 2016). Three different games confirmed the benefits of virtual reality serious games for the improvement of motor functions of stroke patients on cell phones (Song, Ding, Zhao, Jia, & Shull, 2019).

Based on the neuroplasticity phenomenon, movement-based rehabilitation methods in daily activities provoke motor learning, causing permanent changes. Of these methods are constraint-induced movement therapy and robotics-assisted training. The former persuade the patients to use the affected limb without the involvement of the unaffected one in accordance with principles of repetition, intensity, and goal orientation (Wardlaw, Sandercock, Dennis, & Starr, 2003), similar to the approach implemented by the ARMStroke application. In the latter, a mechanical device like an exoskeleton assists the affected limb to perform repetitive movements (Fasoli, Krebs, & Hogan, 2004). However, robotics-assisted systems are not easily accessible because of their cost and prohibition.

Artificial intelligence (AI) has the potential to augment physiotherapy outcomes. AI provides new applications ranging from real time video instructions for detecting pose and joint angles in order to receive optimal feedback toward increasing exercise precision and discipline for motivation and psychotherapeutic aims. In this way, AI enhances supportive framework for physiotherapy in the field of digitizing medicine. It also helps to personalize physiotherapy. Totally, AI integrates cognitive behavioral therapy and virtual reality into precise therapeutic strategies. As this technology is not restrained by geographical location, frequency of exercise could be increased via telemedicine (Davids, Lidströmer, & Ashrafian, 2022). AI brings promising hope to generalize service delivery even to poorer societies by reducing costs and enriching physiotherapy with novel technologies.

OpenPose platform is an example of AI contribution to physical rehabilitation. This platform detects posture and hand gesture and, is useful in sports physiotherapy (Godse et al., 2019). It benefits form cost-effectiveness and easy-to-use feature, which is installed on mobile devices. OpenPose system could be additionally equipped with a chatbot, a natural language processing algorithms that adds psychological component to the rehabilitation program (Ramanandi, 2021). This extra cognitive support augments healing, maintains routines, and stimulates compliance. The option of chatbot is also useful for those who require out-of-hours physiotherapy providing better engage and right services especially in distressed patients. Following up the patients, tracking their progress, and determining any possible non-compliance are also possible by chatbot option of the OpenPose platform(Davids et al., 2022).

Machine learning (ML) is a subfacet of AI that programs a computer or other electronic devices by mathematical instructions in order to perform automated objectives independent of human supervision (Davids et al., 2022). The benefits of ML methods are remarkable escalation in scope and volume of particularly ill patients. This helps physiotherapists for decision-making and assesses patients for best-suited physical therapy (Davids et al., 2022). Based on certain features like duration of in-hospital stay, duration of stroke, and Bartels index score, ML algorithms are designed to predict the scope of recovery after stroke (Ku, Chen, Yang, Lai, & Wang, 2020; Vahlberg & Hellström, 2008). In the context of physiotherapy, AI-based tools like robots or wearable and smart applications facilitate retraining and rehabilitation of patients or simulate workforce education (Davids et al., 2022). However, AI and its branches like ML can not yet replace the physiotherapists since they have a promising role in recovery of cognitive and physical functions in either pre- or post-discharge settings.

Self-training and task-oriented applications with the potential of promoting isolated movements and cognitive tasks are valuable tools in the rehabilitation of stroke patients. However, using applications by stroke patients with hand impairment is debatable. Using fingers in a coordinative and fast manner is defined as a hand skill (Kizony et al., 2016). Aging attenuates hand function (Carmeli, Patish, & Coleman, 2003) and manual dexterity (Desrosiers, Hébert, Bravo, & Rochette, 1999). Hand dexterity in aged individuals, either with or without stroke-related disability, affects working with tablet applications.

Many of these games and applications are motivational. They encourage patients to repeat exercises that are especially useful concerning finger and hand movements. Besides enjoyability, applications should be challenging, which is vital for motor learning. Altogether, these self-training games improve upper extremity deficits after stroke. They could potentially restore some factors like speed of motion, reaction time, and quantification of patients’ performance and progress. Also, it was reported that tablet games help engagement and socialization in patients with subacute stroke (White et al., 2015).

Conclusion

In the present study, novel innovations in the era of mobile applications for the rehabilitation of stroke survivors were presented. In particular, six mobile-based games have been discussed in detail. Each game possesses some merits and limitations. Providers try to present a comprehensive application by which most stroke survivors with different impairments can use it. Such novel remedies give promising rehabilitation opportunities, even for hospitalized patients, where traditional therapy is not feasible. Moreover, they enhance routine rehabilitation programs for survivors. Participants declared that convenience and usability are prominent aspects of such rehabilitation approaches. Future research could be focused on refining the games and implementing extensive studies with a systematic protocol to precisely monitor improvements in different time points.