Abstract
One of the major research fields in medical applications is Computer-aided dementia diagnosis since it progressively declines the cognitive function and afflicts millions of people worldwide, becoming a leading cause of mortality and morbidity of elder people. Pattern recognition methods, applied to dementia diagnosis, improve either the feature extraction or the classifier stage. Particularly, deep learning machines have raised attention to clinical applications since they work in both stages to enhance the system performance. However, the architecture of these machines is highly complex, making hard their training procedures. In this work, we propose a deep supervised feature extraction approach using General Stochastic Networks through a supervised layer-wise non-linear mapping learning. To this end, we maximize the centered kernel alignment function, which accounts for the provided discriminative information regarding the projection of each layer of the network. Our proposal improves the classifier performance by highlighting the class discrimination during the training stage. Besides, we provide a non-linear relevance measure assessing the contribution of the input feature set to build each latent space which is related to the clinical knowledge. Comparison against other automated diagnosis approaches using different features and classification machines is presented for multi-class and bi-class scenarios on the widely-known ADNI database. As a result, our proposal outperforms the compared approaches, reduces the class biasing, and enhances clinical interpretability.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bron, E.E., Smits, M., et al.: Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: the CADDementia challenge. NeuroImage 111, 562–579 (2015)
Sabuncu, M.R., Konukoglu, E.: Clinical prediction from structural brain MRI scans: a large-scale empirical study. Neuroinformatics 13(1), 31–46 (2015)
Sørensen, L., Pai, A., et al.: Dementia diagnosis using MRI cortical thickness, shape, texture, and volumetry. In: Proceedings of MICCAI Workshop Challenge on Computer-Aided Diagnosis of Dementia Based on Structural MRI Data, pp. 111–118 (2014)
Khedher, L., Ramírez, J., Górriz, J.M., Brahim, A., Illán, I.A.: Independent component analysis-based classification of Alzheimer’s disease from segmented MRI data. In: Ferrández Vicente, J., Álvarez-Sánchez, J., de la Paz López, F., Toledo-Moreo, F., Adeli, H. (eds.) IWINAC 2015. LNCS, vol. 9107, pp. 78–87. Springer, Cham (2015). doi:10.1007/978-3-319-18914-7_9
Wachinger, C., Reuter, M., et al.: Domain adaptation for Alzheimer’s disease diagnostics. NeuroImage 139, 470–479 (2016)
Termenon, M., Graña, M., et al.: Brain MRI morphological patterns extraction tool based on extreme learning machine and majority vote classification. Neurocomputing 174, 344–351 (2016)
Liu, S., Liu, S., et al.: Multimodal neuroimaging feature learning for multiclass diagnosis of Alzheimer’s disease. IEEE Trans. Biomed. Eng. 62(4), 1132–1140 (2015)
Suk, H.-I., Lee, S.-W., et al.: Hierarchical feature representation and multimodal fusion with deep learning for AD/MCI diagnosis. NeuroImage 101, 569–582 (2014)
Brockmeier, A., Choi, J., et al.: Neural decoding with kernel-based metric learning. Neural Comput. 26, 1080–1107 (2014)
Buckner, R.L., Head, D., et al.: A unified approach for morphometric and functional data analysis in young, old, and demented adults using automated atlas-based head size normalization: reliability and validation against manual measurement of total intracranial volume. NeuroImage 23(2), 724–738 (2004)
Zöhrer, M., Pernkopf, F.: General stochastic networks for classification. In: Advances in Neural Information Processing Systems, vol. 27, pp. 2015–2023. Curran Associates Inc. (2014)
Zhan, L., Liu, Y., Alzheimer’s Disease Neuroimaging Initiative ADNI, et al.: Boosting brain connectome classification accuracy in Alzheimer’s disease using higher-order singular value decomposition. Front. Neurosci. 9, 257 (2015)
Cárdenas-Peña, D., Collazos-Huertas, D., Castellanos-Dominguez, G.: Centered kernel alignment enhancing neural network pretraining for MRI-based dementia diagnosis. Comput. Math. Methods Med. 2016, 10 p. (2016). doi:10.1155/2016/9523849. Article no. 9523849
Bron, E.E., Smits, M., et al.: Feature selection based on the SVM weight vector for classification of dementia. IEEE J. Biomed. Health Inform. 19(5), 1617–1626 (2015)
Acknowledgments
This work was supported by the research project 111974455497 founded by COLCIENCIAS.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Collazos-Huertas, D., Tobar-Rodriguez, A., Cárdenas-Peña, D., Castellanos-Dominguez, G. (2017). MRI-Based Feature Extraction Using Supervised General Stochastic Networks in Dementia Diagnosis. In: Ferrández Vicente, J., Álvarez-Sánchez, J., de la Paz López, F., Toledo Moreo, J., Adeli, H. (eds) Natural and Artificial Computation for Biomedicine and Neuroscience. IWINAC 2017. Lecture Notes in Computer Science(), vol 10337. Springer, Cham. https://doi.org/10.1007/978-3-319-59740-9_36
Download citation
DOI: https://doi.org/10.1007/978-3-319-59740-9_36
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-59739-3
Online ISBN: 978-3-319-59740-9
eBook Packages: Computer ScienceComputer Science (R0)