Reconstruction of muscle EMG using synergies and artificial neural networks
Evaluation of muscle activity using electromyography (EMG) is important in assessing muscle function during rehabilitation, sports training, and for driving powered prosthetic limbs. However, obtaining EMG data for large numbers of muscles is experimentally difficult. This project aims to quantify the way the central nervous system activates large numbers of muscles using a small set of pre-defined motor control patterns. This strategy will be exploited to efficiently evaluate EMG in all the major upper limb muscles using a low-cost wearable sensor. This approach will also be modelled using machine learning techniques that adopt artificial neural networks. The findings have important implications for smart, personalised medicine.
Leader: David Ackland
Collaborators: Saman Halgamuge (Australian National University)
Convergence of engineering and IT with the life sciences
artificial intelligence; biomechanics; biomedical engineering; biosignals; machine learning