An Instance-Based Algorithm With Auxiliary Similarity Information for the Estimation of Gait Kinematics From Wearable Sensors

• Published in: IEEE transactions on neural networks Vol. 19; no. 9; pp. 1574 - 1582
• Main Authors: Goulermas, J.Y., Findlow, A.H., Nester, C.J., Liatsis, P., Xiao-Jun Zeng, Kenney, L., Tresadern, P., Thies, S.B., Howard, D.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.09.2008; Institute of Electrical and Electronics Engineers
• Subjects: Acceleration; Algorithms; Angular velocity; Anthropometry; Applied sciences; Artificial Intelligence; Auxiliary information; Biomechanical Phenomena - instrumentation; Biomechanical Phenomena - methods; Computer science; control theory; systems; Computer Simulation; Connectionism. Neural networks; Estimates; Exact sciences and technology; Foot; gait; Gait - physiology; generalized regression neural network (GRNN); Hip; Humans; joint kinematics estimation; Kinematics; Knee; Knees; Models, Biological; Models, Theoretical; Monitoring, Ambulatory - instrumentation; Monitoring, Ambulatory - methods; Motion measurement; Neural networks; Neural Networks (Computer); Pattern Recognition, Automated - methods; Similarity; Wearable; Wearable sensors; Human; Legged locomotion; Infrared thermography; Gait; Similarity; Measurement sensor; generalized regression neural network (GRNN); Neural network; Auxiliary information; Adaptive method; Foot; Posture; Osteoarticular system; Walking; Weighting; joint kinematics estimation; Kinematics; Bandwidth; Angular velocity; Medical application
• ISSN: 1045-9227; 1941-0093; 1941-0093
• DOI: 10.1109/TNN.2008.2000808

Wearable human movement measurement systems are increasingly popular as a means of capturing human movement data in real-world situations. Previous work has attempted to estimate segment kinematics during walking from foot acceleration and angular velocity data. In this paper, we propose a novel neural network [ G RNN with A uxiliary S imilarity I nformation (GASI)] that estimates joint kinematics by taking account of proximity and gait trajectory slope information through adaptive weighting. Furthermore, multiple kernel bandwidth parameters are used that can adapt to the local data density. To demonstrate the value of the GASI algorithm, hip, knee, and ankle joint motions are estimated from acceleration and angular velocity data for the foot and shank, collected using commercially available wearable sensors. Reference hip, knee, and ankle kinematic data were obtained using externally mounted reflective markers and infrared cameras for subjects while they walked at different speeds. The results provide further evidence that a neural net approach to the estimation of joint kinematics is feasible and shows promise, but other practical issues must be addressed before this approach is mature enough for clinical implementation. Furthermore, they demonstrate the utility of the new GASI algorithm for making estimates from continuous periodic data that include noise and a significant level of variability.