Learning to Decode Cognitive States from Brain Images
Machine Learning
The support vector decomposition machine
ICML '06 Proceedings of the 23rd international conference on Machine learning
Generalized sparse classifiers for decoding cognitive states in fMRI
MLMI'10 Proceedings of the First international conference on Machine learning in medical imaging
Predicting functional brain ROIs via fiber shape models
MICCAI'11 Proceedings of the 14th international conference on Medical image computing and computer-assisted intervention - Volume Part II
Decoding visual brain states from fMRI using an ensemble of classifiers
Pattern Recognition
| Hi-index | 0.00 |
To understand human brain functioning, task-specific analyses are extensively used. Functional Magnetic Resonance (fMR) images of subjects performing well-defined tasks are utilized. Here, for categorization of distinct cognitive states, a novel scheme that determines the most relevant voxels, using iterative classification, is proposed. In the proposed method, to distinguish between the chosen tasks, baseline classification performance using all active voxels is obtained initially. Subsequently, the brain volume is divided into 4 granules, where voxels belonging to each, are separately used for classification. The best-performing granule is weighted correspondingly higher, in the next iteration. The process of division is continued within the best-performing region. Classification is iteratively carried out till there is no significant change in performance. 10 real scan volumes from 2 public datasets are used to illustrate the performance of the proposed method. The performance of the proposed scheme in distinguishing cognitive tasks considered for the experiment is evaluated to be 99%.