期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2015
卷号：4
期号：9
页码：8269
DOI：10.15680/IJIRSET.2015.0409037
出版社：S&S Publications
摘要：Neural network have impressively demonstrated the capability of modeling spatial information. On theother hand, the application of parallel distributed models to processing of temporal data has been restricted in severalplaces . In space time neural network, the synaptic weights between two artificial neurons are replaced with anadaptable – adjustable filter. Instead of single synaptic weight, it provides a plurality of weights, which represents notonly association but also temporal dependencies. Now the synaptic weights of the neural network are the coefficients ofthe adaptable digital filter, which gives a neural network that is defined both spatially and temporally. This network is aTime-Delayed neural network and also known as Elman network. In this paper, an attempt is made to design a Spatio-Temporal artificial neural network which is used to classify arrhythmias. In this Spatio Temporal Neural Network thespatial information is the output from the 12 leads of ECG, which is obtained from an electrocardiograph, and temporalinformation from a particular lead output. The output of this model is that particular arrhythmia from which patient issuffering.Spatio temporal neural network differs from standard neural network in terms of memory. In STNN thenetworks memory is distributed over the connections rather than being implemented as a special layer of nodes. STNNcan deal with spatial as well as temporal data.
