Data fusion and feature selection for Alzheimer's diagnosis
BI'10 Proceedings of the 2010 international conference on Brain informatics
Computers in Biology and Medicine
| Hi-index | 0.00 |
The purpose of our study was to explore the automatic method of detecting the gray matter loss of Alzheimer’s disease (AD) patients with magnetic resonance imaging (MRI). In this paper, we tried to combine voxel-based morphometry(VBM) with Artificial Neural Network(ANN) in diagnosing Alzheimer’s disease(AD). Firstly, the features were obtained by VBM method, and then principal component analysis(PCA) was used for feature dimensionality reduction to improve its efficiency. Then, a single three-layer, feed-forward ANN with a back-propagation algorithm was used as a classifier. Finally, the performance of the neural network was evaluated by validation protocol. By taking 20 features into account, which were obtained by VBM. The overall predictive capability of Back-Propagation Neural Network in differentiating AD from normal controls reached 100%. The network kept the accuracy of classification consistently. Also, the result will be slightly better with the process of PCA. In conclusion, the results of this study confirmed that the method of combination VBM with ANN could be used in the early diagnosis of AD.