Unsupervised Feature Selection via Adaptive Embedding and Sparse Learning for Parkinson’s Disease Diagnosis
- 679 Downloads
Parkinson’s disease (PD) is known as a progressive neurodegenerative disease in elderly people. Apart from decelerating the disease exacerbation, early and accurate diagnosis also alleviates mental and physical sufferings and provides timely and appropriate medication. In this paper, we propose an unsupervised feature selection method via adaptive manifold embedding and sparse learning exploiting longitudinal multimodal neuroimaging data for classification and regression prediction. Specifically, the proposed method simultaneously carries out feature selection and dynamic local structure learning to obtain the structural information inherent in the neuroimaging data. We conduct extensive experiments on the publicly available Parkinson’s progression markers initiative (PPMI) dataset to validate the proposed method. Our proposed method outperforms other state-of-the-art methods in terms of classification and regression prediction performance.
KeywordsParkinson’s disease Unsupervised feature selection Adaptive manifold embedding Classification Regression prediction
This work was supported partly by the Integration Project of Production Teaching and Research by Guangdong Province and Ministry of Education (No. 2012B091100495), Shenzhen Key Basic Research Project (No. JCYJ20170302153337765), Guangdong Pre-national Project (No. 2014GKXM054), and Guangdong Province Key Laboratory of Popular High Performance Computers (No. 2017B030314073).
- 2.He, X., Cai, D., Niyogi, P.: Laplacian score for feature selection. In: Advances in Neural Information Processing Systems, pp. 507–514 (2006)Google Scholar
- 8.Mohar, B., Alavi, Y., Chartrand, G., Oellermann, O.R.: The Laplacian spectrum of graphs. Graph Theory. Comb. Appl. 2(871–898), 12 (1991)Google Scholar
- 9.Nie, F., Zhu, W., Li, X.: Unsupervised feature selection with structured graph optimization. In: Thirtieth AAAI Conference on Artificial Intelligence (2016)Google Scholar
- 10.Shi, L., Du, L., Shen, Y.D.: Robust spectral learning for unsupervised feature selection. In: 2014 IEEE International Conference on Data Mining, pp. 977–982. IEEE (2014)Google Scholar