Fig. 2From: Resolving the effect of wrist position on myoelectric pattern recognition controlA neural network was trained to predict changes in the mean and variance of each EMG feature, from each channel, for each class as a function of the mean and variance of the feature in a neutral position. The output of each neural network was used to generate data to simulate real data collected from all wrist positions. This simulated data was then used to train an LDA and tested using real dataBack to article page