It is showed that patients with neurological impairments and musculoskeletal injuries can be effectively treated with platform-based robotic rehabilitation devices after their health status is stabilized. With the higher repetitions that robotic devices provide, patients can exercise more, which stimulates neural plasticity in neurological patients in the early stages of their recovery. Once patients can walk better, they can transition to conventional walking to further practice walking on different terrain, improve balance, and correct abnormal gait patterns.

Therefore, platform-based robotic training should be routinely adopted in rehab clinics next to traditional physical therapy. In fact, physiotherapists can use robotic equipment as a multiplier to train more patients. Instead of traditional one-to-one practice, therapists can use robotic devices to treat more patients at the same time. This frees up valuable time for therapists to either train severe patients individually or practice more function-based tasks that require the integration of multiple motor skills. By spending their time on this higher value training activities and letting the robotic devices take over the “heavy” routine tasks, therapists can provide an appropriate level of personalized treatment to their patients and increase their efficiency.

Herein found that the VR-based Rutgers ankle and Hunova seemed to be the most effective robots for rehabilitation. Ankle rehabilitation using a VR-based Rutgers ankle robot has been shown to be effective in rehabilitating patients after stroke and various musculoskeletal ankle injuries [25,29,30,52,53,56]. It is also found that rehabilitation treatment with Hunova is an innovative therapeutic option that can be combined with traditional rehabilitation in patients with Parkinson’s disease [35,36,37], a promising tool for the rehabilitation of stroke patients [39]. For spinal cord injuries (SCI), it can be a useful rehabilitation tool for assessment and training [40,41,42]. In addition, rehabilitation with the Hunova allows measurement of important parameters of static and dynamic stability and can focus on a complex sequence of exercises to restore trunk control and reactive balance after traumatic injury. In the elderly population, the Hunova has the potential to effectively predict fall risk [34]. In patients after traumatic injuries, the Hunova can effectively restore trunk control and reactive balance [40,41,42].

Researchers have studied platform-based robotic rehabilitation in different phases (acute phase, subacute phase, and chronic phase) and at different time points in patients after injuries. Saglia et al. [11], have summarized the rehabilitation protocol for ankle injuries. In the early phase of ankle therapy, the patient can hardly move his foot. Therefore, passive exercises are usually required, during which the movement parameters such as speed, amplitude, and number of repetitions can be determined by the physical therapist. Active exercises can then help the patient regain ROM to move the ankle fully again. Strength training includes both isometric and isotonic exercises. In the final phase of rehabilitation, the patient must perform proprioceptive training (e.g., balance exercises) [11]. Therefore, we believe that an early-stage intervention leads to a faster recovery of the patient than a late-stage intervention. The reason for this is that patients need to do passive exercises in the initial phase. After that, patients need to do active exercises to regain ROM and proprioceptive training (like balance exercises). In contrast, late-stage patients need more exercise sessions to rehabilitate and recover. However, it would be interesting to investigate this in the future through further studies with larger samples.

Bessler et al. reported about 17 adverse events, including tissue-related, musculoskeletal, and physiological adverse events (adverse blood pressure changes) with the use of stationary gait robots (exoskeletons and end-effector) [59,60]. However, for the platform-based robotic devices, we found insufficient literature on adverse events related to long-term use and training. We are not sure and cannot predict what type of adverse events will occur in patients trained with platform-based robots. Future research may provide clues about the adverse effects. This inability to predict adverse events exists because research in the field of robotic rehabilitation is still in its beginning stage.