The new method of the selection of features for the k-NN classifier in the arteriovenous fistula state estimation

In this paper the application of a new method of features selection was presented. Its effects were compared with several other methods of features selection. The study were performed using a data set containing samples of the sound signal emitted by the arteriovenous fistula. The aim was to create a solution with multiclass classification based on the k-NN classifier family allowing for effective and credible assessment of the state of arterial-venous fistula.