Deep learning for neuroimaging: a validation study

Frontiers in Neuroscience

Jessica A Turner¹  Sergey M Plis²  Elena A Allen³  Jeffrey D Long⁴  Henry Jeremy Bockholt⁴  Hans J Johnson⁴  Jane ePaulsen⁴  Devon eHjelm⁵  Vince D Calhoun⁵  Ruslan eSalakhutdinov⁶ 

[1]Georgia State University

[2]The Mind Research Network

[3]The University of Bergen

[4]The University of Iowa

[5]The University of New Mexico

[6]The University of Toronto

关键词: Classification;    Huntington Disease;    fMRI;    MRI;    unsupervised learning;    representation learning;   

DOI  :  10.3389/fnins.2014.00229

来源: DOAJ

【 摘 要 】

Deep learning methods have recently made notable advances in the tasks of classification and representation learning. These tasks are important for brain imaging and neuroscience discovery, making the methods attractive for porting to a neuroimager's toolbox.Success of these methods is, in part, explained by the flexibility of deep learning models. However, this flexibility makes the process of porting to new areas a difficult parameter optimization problem.In this work we demonstrate our results (and feasible parameter ranges) in application of deep learning methods to structural and functional brain imaging data.These methods include deep belief networks and their building block the restricted Boltzmann machine. We also describe a novel constraint-based approach to visualizing high dimensional data. We use it to analyze the effect of parameter choices on data transformations.Our results show that deep learning methods are able to learn physiologically important representations and detect latent relations in neuroimaging data.

【 授权许可】

Unknown