AI Therapist Realizing Expert Verbal Cues for Effective Robot-Assisted Gait Training

Chang, Minsu; Kim, Tae Woo; Beom, Jaewon; Won, Sunjae; Jeon, Doyoung

doi:10.1109/tnsre.2020.3038175

Cited by 10 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The following works provide an overview of the use of expert systems in neurorehabilitation. Most works are devoted to developing expert systems to control robots in rehabilitating upper and lower limbs [9][10][11], which enables patients to self-train to regain control over their bodies after traumatic brain damage. Thus, a patient has to repeat individualized moves of the damaged limb, according to the robotic therapist.…”

Section: Related Workmentioning

confidence: 99%

Expert System for Neurocognitive Rehabilitation Based on the Transfer of the ACE-R to CHC Model Factors

et al. 2022

View full text Add to dashboard Cite

This article focuses on developing an expert system applicable to the area of neurocognitive rehabilitation. The benefit of this interdisciplinary research is to propose an expert system that has been adapted based on real patients’ results from the Addenbrooke’s cognitive examination (ACE-R). One of this research’s main results is a unique proposal to transfer the ACE-R result to the CHC (Cattell–Horn–Carroll) intelligence model. This unique approach enables transforming the CHC model domains according to the modified ACE-R factor analysis, which has never been used before. The expert system inference results allow the automated optimized design of a neurorehabilitation plan to train patients’ cognitive functions according to the CHC model. A set of tasks in 6 difficulty levels (Level 1–Level 6) was proposed for each of the nine CHC model domains. For each patient, the ACE-R results helped determine specific CHC domains to be rehabilitated as well as the starting game level for the rehabilitation within each domain. The proposed expert system has been verified on real data of 705 patients and achieved an average error of 5.94% for all CHC model domains. The proposed system is to be included in the outcomes of the research project of the Technology Agency of the Czech Republic as a verified procedure for healthcare providers.

show abstract

Section: Related Workmentioning

confidence: 99%

Expert System for Neurocognitive Rehabilitation Based on the Transfer of the ACE-R to CHC Model Factors

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The device software allows for the storage of extensive patient and session details, as well as therapy reports [ 2 ]. Lower extremity recovery robots can repeat Gait patterns on behalf of a physiotherapist for extended durations of time and provide clinical outcomes by aiding and adjusting the person's movement [ 29 , 30 ]. Additionally, robotic assistance may encourage repetitive and intense training in a safe setting [ 31 , 32 ].…”

Section: Reviewmentioning

confidence: 99%

Use of Robotics in Gait Rehabilitation Following Stroke: A Review

Warutkar¹,

Dadgal²,

Mangulkar³

2022

Cureus

View full text Add to dashboard Cite

A stroke is an acute disruption of focal or global brain activity that last for a day or leads to death. Most stroke patients have an asymmetric gait, lower-extremity stiffness of the affected (hemiplegia) side, and impaired single stance and weight transfer capacity, restricting their locomotor function. Although between 65% and 85% of individuals can walk alone within six months after a stroke with appropriate surgical/pharmaceutical procedures and rehabilitative therapy, poor walking and cardiac efficiency continue to impede everyday walking for hemiplegia patients. Various methods are used to improve gait in stroke patients. Robotic-assisted gait training (RAGT) is given via a robot system device.Ground exoskeletons, end-effector devices, wearable exoskeletons are three types of rehabilitation robots that have been developed. The HAL (Hybrid Assistive Limb) exoskeleton and RoboGait is also modified gait device to enhance gait. Robotic Neurorehabilitation can be a useful technique for reducing gait impairments and, as a result, increasing the standard of living of post-stroke individuals.

show abstract

“…It is thus questionable if a controller's influence on the overall therapy goal is meaningful and can ever be determined, and it might be a reason why such controllers are rarely found in commercial products today. The discrepancy between conventional clinical practice and research is not new, and researchers started to foster robotic solutions that readopt the manifold interactions with the patient through versatile robotic systems (e.g., ANYexo 2.0 [13]), flexible interfaces for therapist (e.g., emulating the therapist [14]), simulations of multi-modal therapeutic interactions (e.g., imitating therapist's verbal cues [15]), and the design of a modular software framework that allow a supervised assembly of complex, individualized control policies from narrow, specific sub-controllers, i.e., a polymorphic control framework [16]. This framework seeks to reintegrate therapists into the control loop in a supervisory role by allowing them to decide, whenever necessary, which metrics to assess and which robot parameters to adjust.…”

Section: Introductionmentioning

confidence: 99%

Can Robotic Feedback and Adaptation Possibilities match Therapeutic Needs? - An Observational Study

Sommerhalder,

Büchi,

Risch

et al. 2024

Preprint

View full text Add to dashboard Cite

Purpose: An observational study on therapists was executed as a joint project between the Sensory-Motor Systems Lab, ETH Zurich, and the Swiss Paraplegic Centre, Nottwil. The primary goal was to establish a methodology to reasonably tailor current robotic systems to the therapist's preferences in terms of their interaction strategies with the patient. Methods: Therapist's interactions with the patient were recorded, either directly or through a robotic device with extended analysis tools and adaptation possibilities. Experience, operability bias and adaptation confidence of therapists were acquired through questionnaires. Correlation maps were derived to quantify the interaction strategies of the therapists. Results: A total of three distinct interaction strategies emerged based on therapist's personal preferences: Observation of compensatory movement and posture issues caused tactile reaction, issues with robotic settings led to robotic support adaptations, and robotic support adaptations preceded support adaptations. Two strategies emerged based on the exercise type: Mainly direct tactile reactions for reach-goal exercises, and mainly robotic support adaptations for nominal path exercises. The adaptation confidence of therapists strongly depended on the chosen strategies. Conclusion: Robotic systems can be tailored to the therapist's preferences in terms of their interaction strategies by quantifying the therapist's interactions with the robot and the patient. Missing parameters and analysis tools can be found by identifying compensatory strategies, i.e., situations where therapists are forced to swap to direct patient interactions or random parameter adaptations. With this study we presented a method to quantify these strategies with feasible effort.

show abstract

AI Therapist Realizing Expert Verbal Cues for Effective Robot-Assisted Gait Training

Cited by 10 publications

References 36 publications

Expert System for Neurocognitive Rehabilitation Based on the Transfer of the ACE-R to CHC Model Factors

Expert System for Neurocognitive Rehabilitation Based on the Transfer of the ACE-R to CHC Model Factors

Use of Robotics in Gait Rehabilitation Following Stroke: A Review

Can Robotic Feedback and Adaptation Possibilities match Therapeutic Needs? - An Observational Study

Contact Info

Product

Resources

About