The aim of this research is to derive and evaluate an automated identification method that can be applied generically to detect and evaluate the state of body functions using coupled Electromyographic, (EMG) signals the in real time space domain.
The research works from the premise that functions of the body are non-linearly coupled, complex and dynamic. Using this approach, system parameters can be derived to detect and predict changes and limiting factors. This work is focusing on the onset of fatigue during lengthy microsurgical procedures. It is difficult for the individual to assess their relative change in performance over time and this can be crucial as many procedures conclude following a final, precise and critical task.
The project is developing new methodology and algorithms that can be used to analyse and measure the hand tremor and fatigue of surgeons while performing surgery. Complexity method, Principle Component analysis, Empirical Mode Decomposition algorithm and non-linear Auto Regressive with exogenous inputs (NARX) neural networks have been used during development of the system.