An evaluation of volume-based morphometry for prediction of mild cognitive impairment and Alzheimer's disease

Abstract

Voxel-based morphometry from conventional T1-weighted images has proved effective to quantify Alzheimer's disease (AD) related brain atrophy and to enable fairly accurate automated classification of AD patients, mild cognitive impaired patients (MCI) and elderly controls. Little is known, however, about the classification power of volume-based morphometry, where features of interest consist of a few brain structure volumes (e.g. hippocampi, lobes, ventricles) as opposed to hundreds of thousands of voxel-wise gray matter concentrations. In this work, we experimentally evaluate two distinct volume-based morphometry algorithms (FreeSurfer and an in-house algorithm called MorphoBox) for automatic disease classification on a standardized data set from the Alzheimer's Disease Neuroimaging Initiative. Results indicate that both algorithms achieve classification accuracy comparable to the conventional whole-brain voxel-based morphometry pipeline using SPM for AD vs elderly controls and MCI vs controls, and higher accuracy for classification of AD vs MCI and early vs late AD converters, thereby demonstrating the potential of volume-based morphometry to assist diagnosis of mild cognitive impairment and Alzheimer's disease.

Keywords: Alzheimer's disease; Brain morphometry; Classification; Image segmentation; Magnetic resonance imaging; Mild cognitive impairment; Support vector machine.

Fig. A.5

Sketch of the MorphoBox processing pipeline.

Fig. A.6

Example segmentation using MorphoBox: coronal view of an input MPRAGE volume and two distinct overlays of maximum probability tissue labels (CSF, GM, WM) and brain structures (lateral ventricles, central nuclei, hippocampus).

Fig. 1

Linear regression plots for total GM volume estimation on the standardized ADNI dataset using FreeSurfer (left) and MorphoBox (right). Black and red dots represent healthy controls and AD patients, respectively. Dotted lines represent the 10 and 90 percentiles for the controls.

Fig. 2

Linear regression plots for temporal GM volume estimation using FreeSurfer (left) and MorphoBox (right). Black and red dots represent healthy controls and AD patients, respectively. Dotted lines represent the 10 and 90 percentiles for the controls.

Fig. 3

Linear regression plots for hippocampus volume detection on the standardized ADNI dataset using FreeSurfer (left) and MorphoBox (right). Black and red dots represent healthy controls and AD patients, respectively. Dotted lines represent the 10 and 90 percentiles for the controls.

Fig. 4

ROC curves corresponding to Figs. 1–3 for AD detection on the standardized ADNI dataset using, from left to right: total GM, temporal GM, and hippocampus normalized volumes estimated by FreeSurfer and MorphoBox, respectively.