During robot-aided motion rehabilitation training, inappropriate difficulty of the training task usually leads the subject becoming bored or frustrated; therefore, the difficulty of the training task has an important influence on the effectiveness of training. In this study, an adaptive task level strategy is proposed to intelligently serve the subject with a task of suitable difficulty. To make the training task attractive and continuously stimulate the patient’s training enthusiasm, diverse training tasks based on grabbing game with visual feedback are developed. Meanwhile, to further enhance training awareness and inculcate a sense of urgency, a dynamic score feedback method is used in the design of the training tasks. Two types of experiments, functional and clinical rehabilitation experiments, were performed with a healthy adult and two recruited stroke patients, respectively. The experimental results suggest that the proposed adaptive task level strategy and dynamic score feedback method are effective strategies with respect to incentive function and rehabilitation efficacy.
Stroke is a cerebral blood circulation disorder, which is mainly divided into hemorrhagic and ischemic stroke based on the pathogenesis . Based on the report , the global prevalence of stroke was 42.4 million in 2015. In China, approximately 13 million individuals are stroke survivors, and the prevalence is increasing, with 2 million new stroke patients yearly . Overall, the mean age of stroke patients worldwide is increasing, whereas the onset of stroke tends to be younger in China  and Sweden . Due to the lack of effective treatment for the disease, stroke is characterized by high mortality and disability. How to effectively treat stroke and reduce the poor consequences is a common problem in the medical field.
Stroke is referred to as a cerebrovascular accident with a sudden decrease in blood supply to the brain tissue, which may result in brain tissue ischemia and brain cell damage. When the brain nerve cells are damaged, the body functions controlled by these nerve cells are impaired. Stroke treatment and rehabilitation are usually divided into two stages, namely, acute and chronic phases of stroke [4, 5]. During the acute phase, the patient’s corresponding function is restored when the impaired neural connections are recovered within the called sensitive time-limited window. Based on neuroplasticity and compensation of brain function theory, in recent decades, many advanced stroke rehabilitation techniques have been developed and utilized, such as robot-aided device , virtual reality , brain stimulation, and constraint-induced therapy, for the patients in the chronic phase of stroke [8, 9]. To aid the patient in recovering the lost function to the greatest extent, these advanced techniques using unconventional drug therapy for the recovery of the patient’s body functions have received increasing attention from researchers for the past few years.
Stroke may be associated with disabilities for the survivor. The disabilities usually affect the activities of daily living, such as motion ability, walking, speech, and cognition [10, 11]. In clinics, motor deficits are some of the most prevalent symptoms, and 69% of stroke patients have some degree of motion disability of the upper extremity . Fortunately, clinic investigations in both human and animal models demonstrate that massive and intensive motion training can induce cortical changes and reorganization, which construct a relative ability to produce skilled action . Thus, motor function improvements beyond the subacute stage might be induced by rehabilitative therapies. Exercise therapy plays an important role in functional recovery and reconstruction, and it is a popular therapeutic method for stroke rehabilitation. Effectiveness of motion training for motor function improvement has been widely reported . In clinics, motion training is usually conducted by a physiotherapist. The traditional hand-to-hand treatment by a physiotherapist has many disadvantages, such as high labor intensity, low efficiency, and rehabilitation effectiveness varying with the physician. To effectively offer stroke patients with modern technology, all kinds of motion training robots are developed to replace the physiotherapist to offer the patient with designed motion training, which presents the advantages with recording process data, high convenience providing task-oriented practice, and high accuracy in measuring outcomes. In recent years, the rehabilitation robot has become a hot topic in the field of robotics. Robotics is increasingly used in poststroke upper extremity rehabilitation . With regard to the upper extremity rehabilitation robot, studies have greatly contributed to the system mechanism design [15, 16], control method [17, 18], rehabilitation training method , visual feedback, and so on . The aim of developing motion-rehabilitation training robots is to help patients affected with motor disability relearn motion skills based on the experience-dependent neural plasticity with robot-aided motion training. How to stimulate the enthusiasm of the subject to the greatest extent is one of the main considered issues throughout the design of the rehabilitation system. Many training or controlling strategies have been adopted to improve training motivation, such as varied training tasks, vivid visual feedback or virtual reality, friendly interaction, and intelligent control methods [21–23]. However, the training task is usually appointed in advance during the robot-aided motion exercise. Too difficult training tasks will lead the trainer to lose confidence, and too easy training tasks will lead to boredom. Therefore, the level of training tasks during one training session needs to be adjusted based on the training performances. Motion training with matching difficulty level can effectively stimulate the enthusiasm of the subject and make the movement undergo better cooperation.
In this study, the training task based on a game with various levels of difficulty was designed to improve the effectiveness of the rehabilitation training for robot-aided free movement training. Moreover, an adaptive strategy for selection of task level and dynamic visual feedback method were adopted; these interventions can provide the patients with the appropriate training task and motivational visual feedback, which may motivate interest in training and participation awareness.
2. Materials and Methods
2.1. Motion Training Type
In clinics, the motion training type of robot-aided rehabilitation exercise is usually varied with the rehabilitant stage and the state of illness. The motion training types are divided into three modes based on the condition that the robot provides auxiliary force: the passive, aided active, and free motions.
The passive motion training is usually utilized in the early recovery phase where the stroke patient does not present any motion ability, and the movement is fully towed by robot following the predefined trajectory. When the stroke patient possesses a certain active ability but cannot completely overcome gravity, the aided active motion is utilized to arouse the active movement consciousness. During the aided active motion, an appropriate aided force is supplied by the robot to help the subject perform training tasks based on the designed control algorithm. Free motion is used in the stage where the subject can fully overcome gravity. Free movement refers to the movement initiated by the patient himself, and the whole movement process is completely self-initiated by the patient. The end of robot manipulation follows the subject’s hand and does not provide any force or any direction guidance for the movement of the patient. Free motion is fully controlled by the trainer, and the exercise process is actually a coordinated control process of body. The active participation of patients is conducive to accelerating the control of the central nervous system reconstruction on the affected limb. Meanwhile, subjects can freely move according to their wishes, which largely increase their confidence and inspire their motion enthusiasm.
With regard to each motion training type, the training form and control strategy are usually different due to the specific characteristics and goals. To increase interest in training, visual feedback techniques are usually used to design and develop free motion training. This study focuses on free movement training and evaluates the effective training methods using an adaptive training strategy and dynamic visual feedback.
2.2. Rehabilitation System Set-Up
The constructed motion rehabilitation system mainly includes two parts: the hardware and software sections. The four-degree-of-freedom Barrett Whole Arm Manipulator (WAM), which has been widely used as an experimental platform in the medical field, was used as the main platform to construct the upper-limb motion rehabilitation system. The Barrett WAM can be well controlled in joint space, and each joint can be driven by setting the control torque; the position of each rotary joint is measured timely. Additionally, the Barrett WAM was designed with cable-driven technology, presenting outstanding back drivability and safety, which is suitable for an ideal hardware platform for motion training. To monitor the interactive force during rehabilitation, a three-dimensional (3-D) force sensor was developed and installed on the end-effector. An arm-support device was designed and installed to support the impaired limb for stroke patients to perform certain types of motion training. In this investigation, the motion rehabilitation system, which is presented in Figure 1, mainly consists of the WAM manipulator, 3-D force sensor, arm-support device, and controlling PC. More detailed information on the constructed motion rehabilitation system can be acquired from our previous studies [24, 25].
Continue —> Experimental Study on Upper-Limb Rehabilitation Training of Stroke Patients Based on Adaptive Task Level: A Preliminary Study