Rule based realtime motion assessment for rehabilitation exercises

Zhao, Wenbing; Lun, Roanna; Espy, Deborah D.; Reinthal, M. Ann

doi:10.1109/cicare.2014.7007845

Cited by 49 publications

(31 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An earlier version of this work was presented at the 2014 IEEE Symposium on Computational Intelligence in Health and e-Health [5]. This study is supported in part by a Faculty Research Development award, a Graduate Faculty Travel award, and several Undergraduate Summer Research awards, all from the Office of Research, Cleveland State University.…”

Section: Acknowledgmentmentioning

confidence: 99%

“…Finally, this research is extended from our previous work on kinematic modeling [1,5]. In our previous work, a tolerance bound is specified for each parameter in the kinematic rules, and if the observed value for any parameter deviates from the expected value by more than the tolerance amount, the current iteration would be classified as "incorrect".…”

Section: Rule Based Approachesmentioning

confidence: 99%

“…To accommodate human errors and measurement errors, the kinematic rules can be modified by adding a tolerance bound for each parameter that is part of a configuration. Basically, the tolerance-bound-based approach applies the classic logic which categories the execution of an exercise to either correct or wrong, as we have done in our previous work [5].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Realtime Motion Assessment For Rehabilitation Exercises: Integration Of Kinematic Modeling With Fuzzy Inference

Zhao

Lun

Espy

et al. 2014

Journal of Artificial Intelligence and Soft Computing Research

Self Cite

View full text Add to dashboard Cite

This article describes a novel approach to realtime motion assessment for rehabilitation exercises based on the integration of comprehensive kinematic modeling with fuzzy inference. To facilitate the assessment of all important aspects of a rehabilitation exercise, a kinematic model is developed to capture the essential requirements for static poses, dynamic movements, as well as the invariance that must be observed during an exercise. The kinematic model is expressed in terms of a set of kinematic rules. During the actual execution of a rehabilitation exercise, the similarity between the measured motion data and the model is computed in terms of their distances, which are then used as inputs to a fuzzy interference system to derive the overall quality of the execution. The integrated approach provides both a detailed categorical assessment of the overall execution of the exercise and the degree of adherence to individual kinematic rules.

show abstract

Section: Acknowledgmentmentioning

confidence: 99%

Section: Rule Based Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Realtime Motion Assessment For Rehabilitation Exercises: Integration Of Kinematic Modeling With Fuzzy Inference

Zhao

Lun

Espy

et al. 2014

Journal of Artificial Intelligence and Soft Computing Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the rule-based human motion assessment approach that we developed previously [4], [5], [6], we define three types of rules for each activity:…”

Section: B Realtime Human Motion Recognition and Assessmentmentioning

confidence: 99%

“…The methodology is adapted from an approach that we developed previously for rehabilitation exercise monitoring [4], [5], [6].…”

Section: Introductionmentioning

confidence: 99%

Privacy-Aware Human Motion Tracking with Realtime Haptic Feedback

Zhao

Deborah

Reinthal

et al. 2015

2015 IEEE International Conference on Mobile Services

Self Cite

View full text Add to dashboard Cite

In this paper, we describe a system that integrates 3D motion sensors, wearable devices, and smart phones to perform privacy-aware human motion tracking with realtime haptic feedback. The system is designed to enhance safe patient handling and thus reduce the injury rates of health caregivers in nursing homes and hospitals. Due to privacy concern and governmental regulations, it is paramount that only consented caregivers are tracked. Our system satisfies this requirement by employing a registration process for consented caregivers before their activities are monitored. The registration process involves the corroboration of human motion captured in different modalities (computer-vision-based and accelerometer-based). Patient handling activities of caregivers are assessed in realtime using a rule-based approach. Upon detection of unsafe activities, haptic feedback is delivered in realtime to the caregiver via the wearable device worn by the caregiver.

show abstract