Forecasted gait trajectories of children could be used as feedforward input to control lower limb robotic devices, such as exoskeletons and actuated orthotic devices (e.g., Powered Ankle Foot Orthosis—PAFO). Several studies have forecasted healthy gait trajectories, but, to the best of our knowledge, none have forecasted gait trajectories of children with pathological gait yet. These exhibit higher inter- and intra-subject variability compared to typically developing gait of healthy subjects. Pathological trajectories represent the typical gait patterns that rehabilitative exoskeletons and actuated orthoses would target. In this study, we implemented two deep learning models, a Long-Term Short Memory (LSTM) and a Convolutional Neural Network (CNN), to forecast hip, knee, and ankle trajectories in terms of corresponding Euler angles in the pitch, roll, and yaw form for children with neurological disorders, up to 200 ms in the future. The deep learning models implemented in our study are trained on data (available online) from children with neurological disorders collected by Gillette Children’s Speciality Healthcare over the years 1994–2017. The children’s ages range from 4 to 19 years old and the majority of them had cerebral palsy (73%), while the rest were a combination of neurological, developmental, orthopaedic, and genetic disorders (27%). Data were recorded with a motion capture system (VICON) with a sampling frequency of 120 Hz while walking for 15 m. We investigated a total of 35 combinations of input and output time-frames, with window sizes for input vectors ranging from 50–1000 ms, and output vectors from 8.33–200 ms. Results show that LSTMs outperform CNNs, and the gap in performance becomes greater the larger the input and output window sizes are. The maximum difference between the Mean Absolute Errors (MAEs) of the CNN and LSTM networks was 0.91 degrees. Results also show that the input size has no significant influence on mean prediction errors when the output window is 50 ms or smaller. For output window sizes greater than 50 ms, the larger the input window, the lower the error. Overall, we obtained MAEs ranging from 0.095–2.531 degrees for the LSTM network, and from 0.129–2.840 degrees for the CNN. This study establishes the feasibility of forecasting pathological gait trajectories of children which could be integrated with exoskeleton control systems and experimentally explores the characteristics of such intelligent systems under varying input and output window time-frames.
Abstract-A fronthaul design for current and future mobile networks based on the transport of sampled radio signals from/to base station baseband processing units (BBUs) to/from remote radio heads (RRHs), is presented. The design is a pure-Ethernet switched architecture that uses virtual local area network (VLAN) identifiers for the RRHs and flow identifiers for the antenna ports, and is compatible with current standardization definitions. A comprehensive analysis for the limits of the Ethernet fronthaul in terms of the total number of antennas that can be supported is carried out, based on the latency imposed by the Ethernet network. The analysis assumes the transportation of control and management (C&M) and timing information (based on the precision-time protocol, PTP) but is valid for other types of background traffic (for example, that generated by the implementation of different longterm evolution (LTE) functional subdivisions, in a fronthaul with mixed processing). A low-cost testbed using "smart SFP" in-line probes is presented and used to obtain measurements from an Ethernet fronthaul, transporting mixed traffic. The measurements show how background traffic affects hybrid-automatic repeat request (HARQ) retransmissions, and are used to validate the analysis. The effects of contention of PTP packets is discussed and a simple solution to overcome the effects of contention is proposed.
This project has received funding from the Interreg 2 Seas programme 2014-2020 co-funded by the European Regional Development Fund under subsidy contract No 2S05-038 (M.O.T.I.O.N project). Rania Kolaghassi acknowledges the support of studentship through M.O.T.I.O.N, and Mohamad Kenan Al-Hares acknowledges the support of M.O.T.I.O.N. Data used in this work is stored in Kent Academic Repository (https://kar.kent.ac.uk/).
Children with a neurological disorder such as cerebral palsy (CP) severely suffer from a reduced quality of life because of decreasing independence and mobility. Although there is no cure yet, a lower-limb exoskeleton (LLE) has considerable potential to help these children experience better mobility during overground walking. The research in wearable exoskeletons for children with CP is still at an early stage. This paper shows that the number of published papers on LLEs assisting children with CP has significantly increased in recent years; however, no research has been carried out to review these studies systematically. To fill up this research gap, a systematic review from a technical and clinical perspective has been conducted, based on the PRISMA guidelines, under three extended topics associated with "lower limb", "exoskeleton", and "cerebral palsy" in the databases Scopus and Web of Science. After applying several exclusion criteria, seventeen articles focused on fifteen LLEs were included for careful consideration. These studies address some consistent positive evidence on the efficacy of LLEs in improving gait patterns in children with CP. Statistical findings show that knee exoskeletons, brushless DC motors, the hierarchy control architecture, and CP children with spastic diplegia are, respectively, the most common mechanical design, actuator type, control strategy, and clinical characteristics for these LLEs. Clinical studies suggest ankle-foot orthosis as the primary medical solution for most CP gait patterns; nevertheless, only one motorized ankle exoskeleton has been developed. This paper shows that more research and contribution are needed to deal with open challenges in these LLEs.
The version in the Kent Academic Repository may differ from the final published version. Users are advised to check http://kar.kent.ac.uk for the status of the paper. Users should always cite the published version of record.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.