10 February 2023 in Advanced Intelligent Systems
Pneumatic and tendon‐driven actuators are widely used in soft robotic glove design. Tendon‐driven robotic gloves are generally better in controllability, dexterity, and force output, but they are less comfortable than pneumatic ones. Most soft gloves focus on only one actuation mode where either motor‐driven tendon or pump‐driven pneumatic transmission is used. Herein, a double‐acting soft actuator (DASA) that provides both tendon‐driven flexion and pneumatic extension of fingers by a single motor is presented. This is achieved by a smart pulley and bellow system. The kinematic model of the tendon‐driven flexion and the torque model of the fabric‐based pneumatic extension actuator (FPEA) are developed to analyze the DASA performance. The bending angle of the index finger actuated by the tendon and the FPEA extension torque of a joint are characterized by experimental studies. A cycle test of the DASA is conducted 3000 times, demonstrating its high repeatability. A prototype soft glove (68 g) based on the proposed DASA with a control box (835 g) is fabricated to demonstrate finger flexion and extension assistance. Based on electromyography signals, the performance of the robotic glove is evaluated by a squeezing sponge test.
18 January 2022 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Surface electromyogram pattern recognition (EMG-PR) requires time-consuming training and retraining procedures for long-term use, hindering the usability of myoelectric control. In this paper, we design a fabric myoelectric armband to reduce the electrode shifts. Furthermore, we propose a fully unsupervised adaptive approach called hybrid serial classifier (HSC) to eliminate the burden of retraining over multiply days. We investigated the performance of our approach with a dataset of ten types of forearm motion from ten male subjects over eight weeks (total ten days, including: from day 1 to day 7, day 14, day 28, day 56). The average inter-day classification accuracies of HSC without any new retraining data are 86.61% when trained exclusively with the first day’s EMG data, and 94.77% when trained with other nine days’ data. We compare our proposed HSC algorithm with linear discriminant analysis (LDA) without recalibration (BLDA) and supervised adaption LDA (ALDA) with just one trial of new retraining data. The inter-day classification accuracy of HSC is significantly higher than that of BLDA and ALDA. These results indicate that our novel armband sEMG device is feasible for long-term use in conjunction with the proposed HSC algorithm.
24 February 2025 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
As computer science progresses, neuromusculoskeletal models are increasingly applied in clinical settings, particularly when studying abnormal characteristics in patients with cerebral palsy. Digital neuromusculoskeletal models enable researchers and clinicians to gain a deeper understanding of movement mechanisms, providing additional insights for diagnosis and treatment. While biomechanical simulation platforms like OpenSim offer standardized neuromusculoskeletal models for simulation, relying on generic healthy models to simulate movements in cerebral palsy patients can lead to inaccuracies. Therefore, personalized muscle-tendon parameters are essential for cerebral palsy patient models. In this study, we collected ultrasound video data of the semitendinosus muscle from two patients with cerebral palsy during the passive knee extension process. We proposed a muscle-tendon parameter personalization method and developed the individualized OpenSim models for the patients using this data. We validated the personalized models’ output fiber length and pennation angle through a series of hip flexion movement tests. The experimental results demonstrate that using the personalized muscle model for cerebral palsy patients produces muscle fiber length and pennation angle more closely aligned with ultrasound-measured values. After personalization, the RMSE between model output and ultrasound measurement of muscle fiber length and pennation angle decreased by 96.80% and 61.80%, respectively, averaged across both subjects. This study introduces a method for determining muscle-tendon parameters in cerebral palsy patients' digital neuromusculoskeletal models, providing researchers and clinicians with more precise biomechanical information. These insights can better inform the treatment of cerebral palsy patients, ultimately enhancing therapeutic outcomes.
1 January 2026 in IEEE Transactions on Neural Systems and Rehabilitation Engineering