Penalized Likelihood Phenotyping: Unifying Voxelwise Analyses and Multi-Voxel Pattern Analyses in Neuroimaging
- PDF / 1,314,527 Bytes
- 21 Pages / 595.224 x 790.955 pts Page_size
- 47 Downloads / 173 Views
ORIGINAL ARTICLE
Penalized Likelihood Phenotyping: Unifying Voxelwise Analyses and Multi-Voxel Pattern Analyses in Neuroimaging Penalized Likelihood Phenotyping Nagesh Adluru · Bret M. Hanlon · Antoine Lutz · Janet E. Lainhart · Andrew L. Alexander · Richard J. Davidson Published online: 10 February 2013 © Springer Science+Business Media New York 2013
Abstract Neuroimage phenotyping for psychiatric and neurological disorders is performed using voxelwise analyses also known as voxel based analyses or morphometry (VBM). A typical voxelwise analysis treats measurements at each voxel (e.g. fractional anisotropy, gray matter probability) as outcome measures to study the effects of possible explanatory variables (e.g. age, group) in a linear regression setting. Furthermore, each voxel is treated independently until the stage of correction for multiple comparisons. Recently, multi-voxel pattern analyses (MVPA), such as classification, have arisen as an alternative to VBM. The main advantage of MVPA over VBM is that the former employ multivariate methods which can account for interactions among voxels in identifying significant patterns. They also provide ways for computer-aided diagnosis and prognosis at individual subject level. However, compared to VBM, the results of MVPA are often more difficult to interpret and prone to arbitrary conclusions. In this paper, first we use penalized likelihood modeling to provide a unified framework for understanding both VBM and MVPA. We then utilize statistical learning theory to provide practical methods for interpreting the results of MVPA beyond commonly used performance metrics, such as leave-one-out-cross validation accuracy and area under the receiver operating characteristic (ROC) curve. Additionally, we demonstrate that there are challenges in MVPA when trying to obtain image
N. Adluru () · B. M. Hanlon · A. Lutz · A. L. Alexander · R. J. Davidson University of Wisconsin-Madison, Madison, WI, USA e-mail: [email protected] J. E. Lainhart University of Utah, Salt Lake City, UT, USA
phenotyping information in the form of statistical parametric maps (SPMs), which are commonly obtained from VBM, and provide a bootstrap strategy as a potential solution for generating SPMs using MVPA. This technique also allows us to maximize the use of available training data. We illustrate the empirical performance of the proposed framework using two different neuroimaging studies that pose different levels of challenge for classification using MVPA. Keywords Classification · Regression · Voxel based morphometry · Multi-Voxel pattern analysis · Generalization risk · Image phenotyping · Penalized likelihood · Linear models
Introduction Many neuroimaging studies are conducted with a priori hypotheses to be tested. Voxelwise analysis1 (henceforth referred to as VBM) is the most widely used framework for hypothesis testing in neuroimaging. In this framework, the measurements at each voxel (or region) are treated as outcome measures and are analyzed independently leading to a large number of univ
Data Loading...