We propose to develop a prototype device for reliable and robust identification of clinically important muscle parameters out of non-invasively acquired surface electromyograms (SEMG) only.
The project covers all aspects from basic research on compound signal decomposition techniques and SEMG signals, to the design and implementation of SEMG acquisition device. It is also highly multidisciplinary integrating the latest developments in signal processing with the state-of-the-art in neurophysiology.
The main objectives of the project include
- the design and implementation of 2D high-density electrode arrays, suitable for recoding the SEMG signals from different types of muscles,
- the development and implementation of advanced procedures for automatic decomposition of SEMG signals into its constituent motor unit action potential trains,
- validation and evaluation of the decomposition techniques on both synthetic and real SEMG signals, recorded in different conditions from different muscles.
Decomposition techniques will be based on a completely novel blind source separation concept, enabling fast, robust and reliable estimation of signals constituting SEMG. Special care will be taken to enable the decomposition of noisy and underdetermined mixtures (i.e. mixtures with more sources than observations) of not strictly orthogonal sources, thereby significantly extending the limits of the state-of-the-art in blind source separation. The surface electrode arrays will include at least several tens of electrodes arranged in different matrix configurations.
The project is expected to provide a novel, non-invasive and powerful tool for basic and applied research and clinical investigations in occupational, sport and rehabilitation medicine, ergonomics and space medicine, to have a high social and economic impact, and to considerably reduce the cost of monitoring the effectiveness of current motor rehabilitation treatments.
Fields of science
Call for proposal
See other projects for this call