Children with cerebral palsy (CP) have sensorimotor impairments including weakness, spasticity, reduced motor control and sensory deficits. Proprioceptive dysfunction compounds the decreased motor control and mobility. The aims of this paper were to (1) examine proprioceptive deficit of lower extremities of children with CP; (2) study improvement in proprioception and clinical impairments through robotic ankle training (RAT). Eight children with CP participated in a 6-week RAT with pre and post ankle proprioception, clinical, biomechanical assessment compared to the assessment of eight typically developing children (TDC). The children with CP participated in passive stretching (20 min/session) and active movement training (20 to 30 min/session) using an ankle rehabilitation robot (3 sessions/week over 6 weeks, total of 18 sessions). Proprioceptive acuity measured as the plantar and dorsi-flexion motion at which the children recognized the movement was 3.60 ± 2.28° in dorsiflexion and −3.72 ± 2.38° in plantar flexion for the CP group, inferior to that of the TDC group’s 0.94 ± 0.43° in dorsiflexion (p = 0.027) and −0.86 ± 0.48° in plantar flexion (p = 0.012). After training, ankle motor and sensory functions were improved in children with CP, with the dorsiflexion strength increased from 3.61 ± 3.75 Nm to 7.48 ± 2.75 Nm (p = 0.018) and plantar flexion strength increased from −11.89 ± 7.04 Nm to −17.61 ± 6.81 Nm after training (p = 0.043). The dorsiflexion AROM increased from 5.58 ± 13.18° to 15.97 ± 11.21° (p = 0.028). The proprioceptive acuity showed a trend of decline to 3.08 ± 2.07° in dorsiflexion and to −2.59 ± 1.94° in plantar flexion (p > 0.05). The RAT is a promising intervention for children with CP to improve sensorimotor functions of the lower extremities. It provided an interactive and motivating training to engage children with CP in rehabilitation to improve clinical and sensorimotor performance.

The model-based gait analysis of kinematic characteristics of the human body has been used to identify individuals. To extract gait features, spatiotemporal changes of anatomical landmarks of the human body in 3D were preferable. Without special lab settings, 2D images were easily acquired by monocular video cameras in real-world settings. The 2D and 3D locations of key joint positions were estimated by the 2D and 3D pose estimators. Then, the 3D joint positions can be estimated from the 2D image sequences in human gait. Yet, it has been challenging to have the exact gait features of a person due to viewpoint variance and occlusion of body parts in the 2D images. In the study, we conducted a comparative study of two different approaches: feature-based and spatiotemporal-based viewpoint invariant person re-identification using gait patterns. The first method is to use gait features extracted from time-series 3D joint positions to identify an individual. The second method uses a neural network, a Siamese Long Short Term Memory (LSTM) network with the 3D spatiotemporal changes of key joint positions in a gait cycle to classify an individual without extracting gait features. To validate and compare these two methods, we conducted experiments with two open datasets of the MARS and CASIA-A datasets. The results show that the Siamese LSTM outperforms the gait feature-based approaches on the MARS dataset by 20% and 55% on the CASIA-A dataset. The results show that feature-based gait analysis using 2D and 3D pose estimators is premature. As a future study, we suggest developing large-scale human gait datasets and designing accurate 2D and 3D joint position estimators specifically for gait patterns. We expect that the current comparative study and the future work could contribute to rehabilitation study, forensic gait analysis and early detection of neurological disorders.

Accurately identifying human key points is crucial for various applications, including activity recognition, pose estimation, and gait analysis. This study introduces a high-resolution dataset formed via the VICON motion capture system and three diverse 2D cameras. It facilitates the training of neural networks to estimate 2D key joint positions from images and videos. The study involved 25 healthy adults (17 males, 8 females), executing normal gait for 2 to 3 s. The VICON system captured 3D ground truth data, while the three 2D cameras collected images from different perspectives (0°, 45°, and 135°). The dataset was used to train the Body Pose Network (BPNET), a popular neural network model developed by NVIDIA TAO. Additionally, a comparison entails another BPNET model trained on the COCO 2017 dataset, featuring over 118,000 annotated images. Notably, the proposed dataset exhibited a higher level of accuracy (14.5%) than COCO 2017, despite comprising one-fourth of the image count (23,741 annotated image). This substantial reduction in data size translates to improvements in computational efficiency during model training. Furthermore, the unique dataset’s emphasis on gait and precise prediction of key joint positions during normal gait movements distinguish it from existing alternatives. This study has implications ranging from gait-based person identification, and non-invasive concussion detection through sports temporal analysis, to pathologic gait pattern identification.

Composting offers a practical pathway for recycling organic waste, yet maintaining favorable process conditions and translating composting into structured experiential learning remain challenging. This study designed and conducted a preliminary field evaluation of an automated, sensor-enabled composting system intended to process vegetative waste from a small-scale farm while supporting sustainability education. The prototype combined an aerated static-pile configuration with forced aeration, humidity-triggered irrigation, three vertically distributed temperature and relative-humidity sensors, oxygen and fill-level monitoring, a touchscreen interface, transparent viewing panels, multiple access points, and remote data logging. A 29-day trial was conducted in an unheated tunnel greenhouse using approximately 1 m³ of vegetative waste and wood chips mixed at a reported 1:1 mass ratio. Sensor data were recorded at 10-minute intervals and analyzed descriptively to examine temporal patterns, vertical variation, system performance, and data completeness. Temperatures showed recurrent diurnal fluctuations and clear vertical stratification, reaching maximum values of 54.39°C, 45.83°C, and 35.67°C in the upper, middle, and lower zones, respectively. Lower-zone measurements were incomplete because of a connection failure, while middle-zone relative humidity remained near sensor saturation, limiting interpretation. The observed temperature increases, absence of malodor, and visible material changes were consistent with ongoing aerobic decomposition, although compost maturity and educational outcomes were not directly assessed. These findings demonstrate the feasibility of integrating monitoring, user interaction, visual access, and real-time data communication in a smart composter. Future work should validate sensor placement, quantify actuator performance, assess compost quality, and evaluate learning outcomes aligned with the United Nations Sustainable Development Goals.

Background/Objectives: Carpal tunnel syndrome (CTS) is a prevalent neuropathy often associated with repetitive wrist movements, sustained wrist postures, and external compression. Although cycling offers numerous health benefits, prolonged handlebar contact may contribute to hand/wrist discomfort and cycling-related neuropathies. As a pilot study, the aim of this work was intended to (1) evaluate the feasibility of using pressure mapping to quantify glove-related pressure redistribution during cycling; (2) identify key confounding variables requiring improved control in future studies, including comfort/discomfort, posture, cycling experience, hand position, and anthropometric variability; and (3) provide preliminary estimates of pressure variability and effect magnitude between bare hands, conventional biking gloves, and Shock·Tek™ gloves to inform a larger confirmatory study. Methods: Thirteen right-handed adults were tested during low-intensity cycling using a stationary bike equipped with a pliance^®-x pressure sensor wrapped around the handlebar under three conditions: no gloves, traditional padded gloves, and Shock·Tek™ gloves. Results: This exploratory pilot study (1) demonstrated the feasibility of using pressure mapping, (2) identified confounding variables, and (3) showed that while traditional gloves did not significantly reduce surface pressure in the projected median nerve region as we hypothesized, the Shock·Tek™ gloves yielded a 61% reduction in pressure (mean: 20.23 kPa vs. 51.96 kPa with no gloves). Conclusions: These preliminary findings suggest that glove design can influence surface pressure distribution over the projected median nerve region at the hand–glove–handlebar interface during short-duration stationary cycling.