As diabetic neuropathy leads to toe deformities and sensory neuropathy impairs pain perception, using commercially available shoes poses health risks for diabetics. Prescribing appropriate footwear is a challenge; medical professionals need an objective system for evaluating shoe compatibility. This can be achieved by measuring the vibrations (that are generated by the in-shoe vertical load and shear forces) to estimate shear forces. To develop a system that can be used in evaluating shoe compatibility, porcine skin, which is biologically similar to human skin, was used to simulate the human foot in a replication experiment. An ultra-fine piezoelectric wire sensor was used to measure the vibrations that were reproduced. Frequency features were extracted from the vibration data; machine learning was applied to evaluate classification accuracy, and a regression model was constructed to assess estimation accuracy using the root mean square error (RMSE). In this study, the maximum classification accuracy was 93.3% when vertical load was included as a feature after porcine skin was excluded. The RMSE was 0.464 N. Future research will focus on enhancing the accuracy of shear force estimation under real walking conditions by exploring innovative methods for attaching sensors to the bottom and sides of the human foot. This approach will help achieve a more accurate and practical solution for real-time evaluations. Additionally, efforts will be made to improve classification accuracy using only vibration waveforms, leading to a more robust and generalized method for estimating shear forces. By addressing individual variability, future experiments on human feet will incorporate personalized approaches that account for differences in foot anatomy, ensuring broader applicability and reliability. These advancements will contribute to the development of a reliable system for real-time shoe fit evaluation, with the potential to improve comfort and health, especially for individuals with sensory impairments.

The adoption rate of myoelectric prosthetic hands is low owing to the lack of training facilities and challenges associated with prolonged use. A key issue in training is the difficulty users face in adapting to these devices, which we attribute to their inability to accurately recognize errors and areas for improvement. To address this, we developed a feedback method that visually simplifies error identification. Specifically, we used a self-organizing map (SOM) to map acquired muscle vibration data from a high to a low-dimensional space, visualizing the characteristics of each movement to identify the variations in movement that cause misrecognition. Furthermore, to promote accurate movement learning, we provided participants with feedback on movement biases and deficiencies. We used an ultrafine-diameter piezoelectric wire sensor to acquire muscle vibration data for seven different movements, including grasping and palm flexion/extension. The SOM was then used to identify the movements and provide feedback. Consequently, the movement identification rate after feedback improved from 91.8% to 96.1%, indicating that participants could more accurately replicate movements. This approach facilitates better movement awareness, contributing to shorter training times and improved proficiency for prosthetic hand users. Furthermore, the proposed feedback method could be applied to rehabilitation programs and training systems for other assistive devices, enhancing their effectiveness by promoting more accurate motor learning.

Although the world's population is growing, the birth rate is currently declining and is expected to decline, resulting in an ageing population. Electric wheelchairs enhance social activities by expanding the mobility range of older people. However, the proportion of accidents involving electric mobility devices is high among people aged 70 years and above. This is because of decreased sensory and cognitive speeds and motor function. To improve these factors, we focused on surface electromyograms (sEMG) that are output before the start of movement. Using this signal, we used machine learning to estimate the distance and direction of movement of a hand before concluding the control operation of the wheelchair. Multiple features, including frequency and magnitude sum, were determined from the measured sEMG, and two types of machine learning, support vector machine (SVM) and random forest (RF), were used to identify the distance and direction of movement. Three types of movement distances and four types of movement directions were identified. The feature importance was calculated, and those contributing to the distance and direction discrimination were determined and compared for each machine learning method. The discrimination rate using an RF was higher than that using SVM. The results are discussed, focusing on the importance of features for each participant.

A sensor that measures the contact force of an object has different specifications depending on the measurement target and the required accuracy. In particular, when measuring the movement of actual robots and equipment and the movement of humans, it is essential not to affect the movement. Therefore, a light and thin sensor is required. In this study, we developed a circular piezoelectric wire sensor to measure vibration of the human body during movement. To estimate the in-shoe shear force, the shear force generated in the shoe was reproduced in the experiment, followed by the measurement of vibration using the circular sensor. Furthermore, a method for estimating the magnitude of the shear force from the vibration features was developed by applying the analysis results of the generated shear force obtained from an existing shear force sensor and the frequency analysis results of the vibration obtained from this sensor to machine learning. From the experimental results, we showed that it is possible to estimate the generated shear force during motion using the proposed circular sensor.

The authors have developed a cane gait-training machine that enables stroke and paraplegic patients to safely rehabilitate on their own. This training machine is pulled by a wire connected to a harness on the patient's waist; thus, the training machine can follow the patient without the use of hands. However, the machine's ability to follow the walker is an issue. Therefore, we motorised the casters and set them to follow the patient's movements. To cope with the transmission, processing, and mechanical delays that occur in this process, and for predictive control, we used long short-term memory, a machine learning method, which predicted the future waist gait speed from the acceleration measured at multiple sites on the body. In this study, we examined the effects on prediction error of varying the combination of acceleration measurement sites used for learning and the prediction horizon, which is the target time to be predicted. Prediction errors under certain conditions enabled the prediction of each subject's average gait speed with an accuracy of 3-5%. Overall, the prediction error increased with longer prediction horizon but temporarily decreased at 0.4 s. Although prediction is possible using only one site on the lower body, we believe that prediction using multiple sites will reduce the error due to noise.

As ageing populations continue to grow, there is an increasing demand for mobility aids such as wheelchairs. However, conventional wheelchair wheels are designed for level ground and are not suitable for travelling over rough terrain or steps. In addition, with the rapid development of mega-cities, more obstacles such as steps and Braille blocks have been introduced, making outdoor mobility even more challenging for wheelchair users. Although electric wheelchairs have been developed for use in various environments, they are costly and not accessible for all users. To address this issue, the authors focused on developing an affordable wheel that can be used in various driving environments, including flat ground, steps, and rough terrain. They designed a wheel that utilizes a slider-crank mechanism, which can be adjusted by the wheelchair user using a central disc. In this paper, the design of the wheel is presented, and its performance was evaluated through experiments involving step climbing and rough terrain simulations. The results showed a reduction in the force required to overcome steps, improved stability when navigating steps, and enhanced driving performance on rough terrain after deploying the new wheel.

This research aims to develop a human machine interface for patients with absolute rest, which exploits the contact force of the fingertip. The interface outputs the magnitude and direction of the contact force, based on the change of the nail color extracted from images with a camera. This interface can determine the force direction in the four ones: forward, back, left, and right. An experiment to operate the mouse cursor on the computer with this interface and track pad shows this interface can accurately operate the mouse cursor than track pad with the same range of fingertip movement.