Publication

Abstract : This study presents an artificial neural network (ANN) approach for the classification of sit to stand (STS) phases. In the first part of the study sEMG, trunk and knee flexion/extension data were acquired from 5 healthy participants performing STS task. In the second part of the study two multilayer perceptron (MLP) models were developed and compared based upon their classification accuracies. Model constituting multimodal input was found to have better classification accuracy in detecting the human intention to perform STS task compared to the other model which was based solely on multi-channel sEMG as input.