7th Joint Seminar on Rehabilitation Engineering and Assistive Technology
Neural Electronic Engineering Laboratory Division of Biomedical Engineering for Health and Welfare Graduate School of Biomedical Engineering Tohoku University Sendai, Japan
Biocybernetics Laboratory Department of Biomedical Engineering Institut Teknologi Sepuluh Nopember (ITS) Surabaya, Indonesia
Schedule: Date: December 13, 2023 Time: 13.00 Surabaya /15.00 Sendai Venue: Online (Zoom Meeting)
Presentation list: 1.) Ryo Kawashima and Takashi Watanabe A Feasibility Study of Prediction of Lower Limb Segment Inclination Angles under Different Walking Speeds
2.) Nabila Alya Rahma, Achmad Arifin, Norma Hermawan Design of Wheelchair Navigation System with Object Detection Using Computer Vision for People with Visual Impairment
3.) Ryunosuke Sato and Takashi Watanabe A Basic Study on Elimination Method of Artifact in EMG Signals for Voluntary Activity Estimation in FES Assist Cycling with Pedaling Wheelchair
4.) Natasya Adinda Weninggalih, Achmad Arifin, Josaphat Pramudijanto Development of Muscle Spasticity Monitoring System in Post-Stroke Rehabilitation with Functional Electrical Stimulation using Fuzzy Decision Supoort System
A Feasibility Study of Prediction of Lower Limb Segment Inclination Angles under Different Walking Speeds
Ryo Kawashima*, Takashi Watanabe**
*Graduate School of Engineering, Tohoku University, Japan
**Graduate School of Biomedical Engineering, Tohoku University, Japan
Abstract
Functional Electrical Stimulation (FES) can be useful to restore walking movement of patients with stroke. However, in current clinical application, open-loop control system is used because electrically stimulated musculoskeletal systems have the delay which makes closed-loop control difficult. The purpose of this study is to develop a method of predicting future lower limb segment inclination angles (foot, shank, and thigh) during walking. The predicted angles would enable the user to walk more freely and reproduce the gait of a normal person by improving the delay in closed-loop FES control. Previous studies on predicting the lower limb segmental inclination angles have only verified it for level walking at a constant walking speed. In this report, we examined the feasibility of predicting future lower limb segment inclination angles under different walking speeds. Five types of healthy walking (Slow, Moderate, Fast, Speed up, and Speed down) were measured with inertial motion unit (IMU) attached to the foot, shank and thigh. For each segment, Long Short-Term Memory (LSTM) network was used to create a model that predicts the angle for 500 ms in the future by inputting the current and past angles of 1.5s. Three types of constant speed walking (Slow, Moderate, and Fast) were used for the training data, and 2 types of walking (Speed up and Speed down) were used for the test data. The predicted angles showed high correlation and similar wave form with those measured angles in Speed up and Speed down walking. The results suggest that it would be feasible to predict the lower limb segment inclination angles under different walking speeds by training with various speed walking data.
Design of Wheelchair Navigation System with Object Detection Using Computer Vision for People with Visual Impairment
Nabila Alya Rahma, Achmad Arifin, Norma Hermawan
Biomedical Engineering Department
Institut Teknologi Sepuluh Nopember (ITS), Surabaya, Indonesia
Persons with motor disability of lower limbs face handicap in locomotion. Electric wheelchair is a mobility device that designed to assist the mobility disabled persons. Such individuals with intact upper limbs can control the electric wheelchair navigation using joystick. However, for mobility disabled persons who also have visual impairment, such an assistive navigation system to enhance mobility and reduce collisions. In this study proposed a system to detect objects or obstacles using a portable camera with computer vision program and ultrasonic sensors. The trained Tensorflow model achieves 85.96% accuracy based on mean Average Precision (mAP) in detecting obstacles. The designed system reduced collision rate from 54% (subject use cane as manual obstacle detector) to 12%. This current step of development result is expected to improve success rate in avoiding collisions.
A Basic Study on Elimination Method of Artifact in EMG Signals for Voluntary Activity Estimation in FES Assist Cycling with Pedaling Wheelchair
Ryunosuke Sato*, Takashi Watanabe**
The pedaling wheelchair, “COGY”, can be used for rehabilitation by a paraplegic patient. While the “COGY” has the advantage that there is little risk of falling and daily use is directly related to rehabilitation, there is still a problem that the wheelchair cannot run stably depending on the severity of the paralysis. In this study, we propose FES-assisted cycling, in which running is assisted by functional electrical stimulation (FES). Voluntary movement levels can be obtained from electromyogram (EMG) signals during FES application and used as feedback for FES assisted cycling control, but the problem is the noise caused by electrical stimulation pulses such as artifacts and M waves. This study aimed to determine a method to remove the noise for estimating the voluntary movement level during FES assist. In this report, two experiments were conducted to compare three different denoising methods: Comb filter, Blanking-Comb filter, and Dual-channel EMG patio-temporal differential (DESTD) Method. In the first experiment, EMG signals during FES were simulated by adding noise to the volitional EMGs, and the voluntary movement levels estimated from the raw volitional EMGs and the simulated EMGs were compared in terms of RMSE values. In the second experiment, we compared the correlation coefficients between the estimated voluntary EMG level and the actually exerted force, and evaluated how well the voluntary EMG level was estimated. The first experiment showed that the Blanking-Comb filter had the lowest RMSE value, indicating its high removal performance, while the second experiment showed that the three removal methods had high correlation coefficients and were able to estimate the voluntary movement level with high accuracy. Based on these results, we conclude that the Blanking-Comb filter would be useful for noise reduction.
Development of Muscle Spasticity Monitoring System in Post-Stroke Rehabilitation with Functional Electrical Stimulation using Fuzzy Decision Support System
Natasya Adinda Weninggalih, Achmad Arifin, Josaphat Pramudijanto
Stroke is a leading cause of death and significant disability. Spasticity, characterized by increased muscle tone that can hinder post-stroke recovery and rehabilitation, occurs in 30%-80% of cases. Functional Electrical Stimulation (FES) therapy has been proven effective in reducing spasticity levels and improving motor function. However, objective spasticity monitoring is needed to evaluate rehabilitation progress. This research is preliminary development of a spasticity monitoring system using the Modified Ashworth Scale (MAS) combined with a biomechanical approach using joint angle and torque data. A Fuzzy Clinical Decision Support System (FCDSS) computational system was implemented to generate MAS scores based on quantitative spasticity data. The results show testing under different simulated spasticity conditions yields results consistent with the expected severity levels. In post-stroke rehabilitation with FES, this system is expected to aid in intensive rehabilitation sessions and monitors spasticity development as an indicator of rehabilitation progress in reducing spasticity. The monitoring system is expected to be effective in post-stroke rehabilitation. With its ability to objectively measure and monitor spasticity levels, this system offers valuable support for medical professionals in monitoring and evaluating patient progress, as well as making necessary adjustments in the rehabilitation program.
Next step of this research is experimental measurement with paralyzed subjects during rehabilitation sessions.
7th Joint Seminar on Rehabilitation Engineering and Assistive Technology Neural Electronic Engineering Laboratory Division of Biomedical Engineering for Health
Departemen Teknik Biomedik akan menyelenggarakan rangkaian acara kuliah tamu “Anatomi dan Fisiologi” Semester Gasal Tahun Ajaran 2023-2024 dengan agenda