Classifying cognitive states from fMRI data using neural networks

• Published in: 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541) Vol. 4; pp. 2871 - 2875 vol.4
• Main Authors: Onut, I.-V., Ghorbani, A.A.
• Format: Conference Proceeding
• Language: English
• Published: Piscataway NJ IEEE 2004
• Subjects: Applied sciences; Artificial intelligence; Biological neural networks; Computer science; Computer science; control theory; systems; Data mining; Electronic mail; Exact sciences and technology; Humans; Magnetic resonance imaging; Neural networks; Niobium; Pattern recognition. Digital image processing. Computational geometry; Support vector machine classification; Support vector machines; Human; Backpropagation; Brain; Central nervous system; Functional analysis; Neural network; Nuclear magnetic resonance imaging; Encephalon; Aggregate model; Information use; Cognitive theory; Backpropagation algorithm; Human activity; Voxel; Classification; Medical imagery; Tridimensional image; Feedforward; Data gathering
• ISBN: 0780383591; 9780780383593
• ISSN: 1098-7576
• DOI: 10.1109/IJCNN.2004.1381114

Since the discovery of functional magnetic resonance imaging (fMRI) studies have proved that this technique is one of the best for collecting vast quantities of data about activity of the human brain. Our aim is to use this information in order to predict the cognitive status of the subject given its fMRI activity. We present a new approach for creating single-subject classifiers using bagging from a pool of feed-forward backpropagation networks. Our experiments indicate that as the number of selected features (voxels) increases, the accuracy of the system increases too. Nevertheless, when the number of voxels exceeds 120, the accuracy of the system rapidly increases from 45% to 70%. Eventually it reaches a (near) saturation point after which the increase in the accuracy is very slow.