Spiking Neural Network Learning Model for Decision Support in Orthodontic Extraction

This paper proposes a supervised delay-adaptation spiking neural network model to support decision making in orthodontic extraction. This temporal coding spiking neural network model employs a Hebbian-based rule to shift the connection delays instead of previous approaches to delay selection. Here, the tuned delays compensate for differences in the input firing times of temporal patterns and enable them to coincide. The coincidence detection capability of the spiking neuron has been utilised for detecting and classifying patterns. The structure of the network is similar to that of an LVQ network except that the output layer neurons are coincidence-detecting spiking neurons. An input pattern is represented by the group of the neuron that is the first to fire among all the competing spiking neurons. The proposed spiking neural network has been utilised to classify orthodontic data. The trained network obtained an average classification accuracy of 89.8% on previously unseen test data. This was achieved with a network of 2x4 spiking neurons trained for 40 epochs using 100 training examples. The classification accuracy of the proposed model was found to be better than that of a multilayer perceptron (MLP) network trained using the error back-propagation algorithm.