International Journal on Advanced Science, Engineering and Information Technology, Vol. 8 (2018) No. 6, DOI:10.18517/ijaseit.8.6.5011

Spatio-Temporal fMRI Data in the Spiking Neural Network

Shaznoor Shakira Saharuddin, Norhanifah Murli

Abstract

Deep learning machine based on Spiking Neural Network (SNN) is currently one of the many techniques in computational intelligence to discover knowledge from the data from various fields.  It has been applied in many application areas which include health, engineering, finances, environment and others.  This paper addresses a classification problem based on a case study of functional Magnetic Resonance Image (fMRI) brain data of a subject Reading a Sentence or Looking a Picture.   In the experiment, the most relevant features (voxels) are selected using Signal to Noise Ratio (SNR) before it is propagated in a SNN-based learning architecture.  The spatio-temporal relationships between Spatio Temporal Brain Data (STBD) are learned and classified accordingly. All brain regions are taken from the 04847 subject. The overall result of the experiment shows that the SNR method produced higher accuracy in classifying data of Reading a Sentence instead of data Looking a Picture.

Keywords:

NeuCube; Feature Selection; Brain Data Classification; Functional Magnetic Resonance Imaging (fMRI); Spatio-Temporal Brain Data (STBD); Signal to Noise Ratio (SNR);

Viewed: 23 times (since Sept 4, 2017)

cite this paper