The GTM classifier and its application to the classification of motor unit action potentials
Adriano de Oliveira Andrade, Slawomir J. Nasuto, Peter Kyberd · CentAUR (University of Reading) · 2005
In this paper we devise a procedure, which we call GTM classifier, for data classification based on the Generative Topographic Mapping (GTM) and apply it to the classification of motor unit action potentials (MUAPs). The results of the classification of experimental MUAPs show that classification success rate of up to 93% may be obtained. They also indicate that the GTM classifier may be successfully employed as a tool for outlier rejection, with rejection rate of up to 100%.