A hybrid neuro-fuzzy system for neuromuscular disorders diagnosis

Motor unit action potentials (MUAPs) recorded during routine electromyography (EMG) examination provide important information for the assessment of neuromuscular disorders. The conventional computer-aid methods of MUAP diagnosis are mainly based on single feature set and single neural network model, and the diagnosis accuracy of which is not always satisfactory. In order to utilize multiple feature sets to improve diagnosis accuracy, a hybrid decision support system based on fusion multiple neural network outputs is presented. Back-propagation (BP) neural networks are used as single diagnosis models in every feature set, i.e. i) time domain measures, ii) autoregressive coefficients, and iii) cepstral coefficients. Then these outputs are combined by fuzzy integral. More excellent diagnosis yield indicates the potential of the system for neuromuscular disorders diagnosis.