School of Engineering researchers have published an article in the journal Sensors, which develops deep learning models to forecast pathological gait trajectories of children which can be integrated with lower limb robotic devices.
The article, entitled ‘Performance of Deep Learning Models in Forecasting Gait Trajectories of Children with Neurological Disorders’, was co-written by Rania Kolaghassi, PhD Electronic Engineering candidate, Dr Mohamad Kenan Al-Hares Research Associate for the MOTION Project, Dr Gianluca Marcelli, Senior Lecturer in Engineering and Dr Konstantinos Sirlantzis, Senior Lecturer in Intelligent Systems.
Forecasted gait trajectories of children could be used as feedforward input to control lower limb robotic devices, such as exoskeletons and actuated orthotic devices. Several studies have forecasted healthy gait trajectories, but 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, the researchers 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. 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.
The full article is available to read on the publisher’s website, here: