Sparse Multimodal Manifold-Regularized Transfer Learning for MCI Conversion Prediction
Effective prediction of conversion of mild cognitive impairment (MCI) to Alzheimer’s disease (AD) is important for early diagnosis of AD, as well as for evaluating AD risk pre-symptomatically. Different from most traditional methods for MCI conversion prediction, in this paper, we propose a novel sparse multimodal manifold-regularized transfer learning classification (SM2TLC) method, which can simultaneously use other related classification tasks (e.g., AD vs. normal controls (NC) classification) and also the unlabeled data for improving the MCI conversion prediction. Our proposed method includes two key components: (1) a criterion based on the maximum mean discrepancy (MMD) for eliminating the negative effect related to the distribution differences between the auxiliary (i.e., AD/NC) and the target (i.e., MCI converters/MCI non-converters) domains, and (2) a sparse semisupervised manifold-regularized least squares classification method for utilization of unlabeled data. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database show that the proposed method can significantly improve the classification performance between MCI converters and MCI non-converters, compared with the state-of-the-art methods.
Unable to display preview. Download preview PDF.
- 2.Zhang, D., Shen, D., Alzheimer’s Disease Neuroimaging Initiative: Multi-modal multi-task learning for joint prediction of multiple regression and classification variables in Alzheimer’s disease. NeuroImage 59, 895-907 (2012)Google Scholar
- 4.Davatzikos, C., Bhatt, P., Shaw, L.M., Batmanghelich, K.N., Trojanowski, J.Q.: Prediction of MCI to AD conversion, via MRI, CSF biomarkers, and pattern classification. Neurobiology of Aging 32, 2322.e19–2322.e27 (2011)Google Scholar
- 5.Duchesne, S., Mouiha, A.: Morphological factor estimation via high-dimensional reduction: prediction of MCI conversion to probable AD. International Journal of Alzheimer’s Disease 2011, 914085 (2011)Google Scholar
- 6.Cuingnet, R., Gerardin, E., Tessieras, J., Auzias, G., Lehericy, S., Habert, M.O., Chupin, M., Benali, H., Colliot, O.: Automatic classification of patients with Alzheimer’s disease from structural MRI: a comparison of ten methods using the ADNI database. NeuroImage 56, 766–781 (2011)CrossRefGoogle Scholar
- 9.Borgwardt, K.M., Gretton, A., Rasch, M.J., Kriegel, H.P., Scholkopf, B., Smola, A.J.: Integrating structured biological data by kernel maximum mean discrepancy. Bioinformatics 22, e49–e57 (2006)Google Scholar
- 10.Chen, X., Pan, W., Kwok, J.T., Carbonell, J.G.: Accelerated gradient method for multi-task sparse learning problem. In: Proceedings Ninth IEEE International Conference on Data Mining, ICDM 2009, pp. 746–751 (2009)Google Scholar
- 14.Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines (2001)Google Scholar