High-dimensional pattern regression using machine learning: From medical images to continuous clinical variables

Abstract

This paper presents a general methodology for high-dimensional pattern regression on medical images via machine learning techniques. Compared with pattern classification studies, pattern regression considers the problem of estimating continuous rather than categorical variables, and can be more challenging. It is also clinically important, since it can be used to estimate disease stage and predict clinical progression from images. In this work, adaptive regional feature extraction approach is used along with other common feature extraction methods, and feature selection technique is adopted to produce a small number of discriminative features for optimal regression performance. Then the Relevance Vector Machine (RVM) is used to build regression models based on selected features. To get stable regression models from limited training samples, a bagging framework is adopted to build ensemble basis regressors derived from multiple bootstrap training samples, and thus to alleviate the effects of outliers as well as facilitate the optimal model parameter selection. Finally, this regression scheme is tested on simulated data and real data via cross-validation. Experimental results demonstrate that this regression scheme achieves higher estimation accuracy and better generalizing ability than Support Vector Regression (SVR).

Imaging predictors of remission to anti-depressant medications in major depressive disorder

“…NBS offers a powerful and complementary approach that can be used to characterize the specific features of a network, while controlling properly for the familywise error (Zalesky et al, 2010). Machine-based learning (support vector machine, SVM and relevance vector machine, RVM) are techniques of supervised classification developed in the field of machine learning which is able look for patterns between independent types of imaging data, robustly separate groups in this multi-dimensional space (Wang et al, 2010). SVM has been successfully applied to imaging data from many psychiatric disorders, including depression (Mwangi et al, 2012).…”

Machine learning algorithms to classify spinal muscular atrophy subtypes

“…One approach for doing this is through support vector regression 30,31 or the more recently described relevance vector machines. 32 The enhanced power of the applied algorithm is probably due to the different muscle characteristics extracted by the 2 modalities being used (QMU and EIM). Although elements of these modalities are somewhat correlated, with a coefficient of determination (R 2 ) of 0.36, the differences in underlying principles ensure that each modality brings new information to the classification process.…”

Morphological covariance in anatomical MRI scans can identify discrete neural pathways in the brain and their disturbances in persons with neuropsychiatric disorders

“…Diffusion Tensor Imaging (DTI) provides the unique opportunity to track fiber pathways in vivo (Bammer et al, 2002;Dotson et al, 2009b;Wang et al, 2010) by quantifying the spatial diffusion of water molecules, thereby allowing inferences about the local properties of brain tissue, especially the integrity and coherence of fiber pathways within white matter (Basser, 1995;Basser and Jones, 2002;Basser et al, 1994;Mori and van Zijl, 2002;Mori and Zhang, 2006;Ou and Davatzikos, 2009;Thambisetty et al, 2012;Zacharaki et al, 2011;Zhang and Davatzikos, 2011). Disturbances in DTI measures within white matter fiber pathways have been associated with various disease processes (Batmanghelich et al, 2012;Casanova et al, 2011;Davatzikos et al, 2008Davatzikos et al, , 2011Haris et al, 2011;Koutsouleris et al, 2012;Kubicki et al, 2002;Lim and Helpern, 2002).…”