Beyond Human or Robot Administered Treadmill Training

Krebs, Hermano Igo; Michmizos, Konstantinos P.; Susko, Tyler; Lee, Hyunglae; Roy, Abhishek; Hogan, Neville

doi:10.1007/978-3-319-28603-7_20

Cited by 5 publications

(3 citation statements)

References 131 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In that sense, the knowledge of CE aims to complement the robot-derived quantitative metrics of motor performance [26], [27], already corroborating that successful motor recovery not only requires repetitive movements [28], but "repetition without repetition" [29] signifying "active participation" during therapy [4], [5]. Indeed, a patient highly engaged in a challenging rehabilitation environment is likely to participate actively in the motor task, which is known to enhance motor learning and thus improve clinical outcomes [8].…”

Section: Discussionmentioning

confidence: 77%

Machine Learning for Motor Learning: EEG-based Continuous Assessment of Cognitive Engagement for Adaptive Rehabilitation Robots

Kumar¹,

Michmizos²

2020

Preprint

View full text Add to dashboard Cite

Although cognitive engagement (CE) is crucial for motor learning, it remains underutilized in rehabilitation robots, partly because its assessment currently relies on subjective and gross measurements taken intermittently. Here, we propose an end-to-end computational framework that assesses CE in real-time, using electroencephalography (EEG) signals as objective measurements. The framework consists of i) a deep convolutional neural network (CNN) that extracts taskdiscriminative spatiotemporal EEG to predict the level of CE for two classes-cognitively engaged vs. disengaged; and ii) a novel sliding window method that predicts continuous levels of CE in real-time. We evaluated our framework on 8 subjects using an in-house Go/No-Go experiment that adapted its gameplay parameters to induce cognitive fatigue. The proposed CNN had an average leave-one-out accuracy of 88.13%. The CE prediction correlated well with a commonly used behavioral metric based on self-reports taken every 5 minutes (ρ=0.93). Our results objectify CE in real-time and pave the way for using CE as a rehabilitation parameter for tailoring robotic therapy to each patient's needs and skills.

show abstract

Section: Discussionmentioning

confidence: 77%

Machine Learning for Motor Learning: EEG-based Continuous Assessment of Cognitive Engagement for Adaptive Rehabilitation Robots

Kumar¹,

Michmizos²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Since a successful motor recovery requires both repetitive movements [38] and active participation [39], an adaptation based on specific movement components could further tailor rehabilitation and promote plasticity. For example, an adaptive robotic therapy could target each one of the two movement components: for movement intent, the robot's assistance would adapt to encourage voluntary movements; for RT, the robot would use the percentage of fast RT as an indicator of functional motor recovery or even relate it to motor skill consolidation.…”

Section: Discussionmentioning

confidence: 99%

Deep Learning of Movement Intent and Reaction Time for EEG-informed Adaptation of Rehabilitation Robots

Kumar

Michmizos

2020

2020 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob)

View full text Add to dashboard Cite

Mounting evidence suggests that adaptation is a crucial mechanism for rehabilitation robots in promoting motor learning. Yet, it is commonly based on robot-derived movement kinematics, which is a rather subjective measurement of performance, especially in the presence of a sensorimotor impairment. Here, we propose a deep convolutional neural network (CNN) that uses electroencephalography (EEG) as an objective measurement of two kinematics components that are typically used to assess motor learning and thereby adaptation: i) the intent to initiate a goal-directed movement, and ii) the reaction time (RT) of that movement. We evaluated our CNN on data acquired from an in-house experiment where 13 subjects moved a rehabilitation robotic arm in four directions on a plane, in response to visual stimuli. Our CNN achieved average test accuracies of 80.08% and 79.82% in a binary classification of the intent (intent vs. no intent) and RT (slow vs. fast), respectively. Our results demonstrate how individual movement components implicated in distinct types of motor learning can be predicted from synchronized EEG data acquired before the start of the movement. Our approach can, therefore, inform robotic adaptation in real-time and has the potential to further improve one's ability to perform the rehabilitation task.

show abstract

“…A recent study pointed out that the robot seemed to be effective only for rehabilitation of patients with mild PD. 56) Krebs et al 57) mentioned that robotic therapy was limited by its incompatibility with human motor neuroscience. Another limitation of the robot is the high costs of development, maintenance, and usage.…”

Section: Application Of the Robot Assistant System In Pdmentioning

confidence: 99%

Can the Latest Computerized Technologies Revolutionize Conventional Assessment Tools and Therapies for a Neurological Disease? The Example of Parkinson’s Disease

Asakawa

Sugiyama

Nozaki

et al. 2019

Neurol. Med. Chir.(Tokyo)

View full text Add to dashboard Cite

Dramatic breakthroughs in the treatment and assessment of neurological diseases are lacking. We believe that conventional methods have several limitations. Computerized technologies, including virtual reality, augmented reality, and robot assistant systems, are advancing at a rapid pace. In this study, we used Parkinson’s disease (PD) as an example to elucidate how the latest computerized technologies can improve the diagnosis and treatment of neurological diseases. Dopaminergic medication and deep brain stimulation remain the most effective interventions for treating PD. Subjective scales, such as the Unified Parkinson’s Disease Rating Scale and the Hoehn and Yahr stage, are still the most widely used assessments. Wearable sensors, virtual reality, augmented reality, and robot assistant systems are increasingly being used for evaluation of patients with PD. The use of such computerized technologies can result in safe, objective, real-time behavioral assessments. Our experiences and understanding of PD have led us to believe that such technologies can provide real-time assessment, which will revolutionize the traditional assessment and treatment of PD. New technologies are desired that can revolutionize PD treatment and facilitate real-time adjustment of treatment based on motor fluctuations, such as telediagnosis systems and “smart treatment systems.” The use of these technologies will substantially improve both the assessment and the treatment of neurological diseases before next-generation treatments, such as stem cell and genetic therapy, and next-generation assessments, can be clinically practiced, although the current level of artificial intelligence cannot replace the role of clinicians.

show abstract

Beyond Human or Robot Administered Treadmill Training

Cited by 5 publications

References 131 publications

Machine Learning for Motor Learning: EEG-based Continuous Assessment of Cognitive Engagement for Adaptive Rehabilitation Robots

Machine Learning for Motor Learning: EEG-based Continuous Assessment of Cognitive Engagement for Adaptive Rehabilitation Robots

Deep Learning of Movement Intent and Reaction Time for EEG-informed Adaptation of Rehabilitation Robots

Can the Latest Computerized Technologies Revolutionize Conventional Assessment Tools and Therapies for a Neurological Disease? The Example of Parkinson’s Disease

Contact Info

Product

Resources

About