A large-scale comparison of cortical thickness and volume methods for measuring Alzheimer's disease severity

Abstract

Alzheimer's disease (AD) researchers commonly use MRI as a quantitative measure of disease severity. Historically, hippocampal volume has been favored. Recently, "AD signature" measurements of gray matter (GM) volumes or cortical thicknesses have gained attention. Here, we systematically evaluate multiple thickness- and volume-based candidate-methods side-by-side, built using the popular FreeSurfer, SPM, and ANTs packages, according to the following criteria: (a) ability to separate clinically normal individuals from those with AD; (b) (extent of) correlation with head size, a nuisance covariatel (c) reliability on repeated scans; and (d) correlation with Braak neurofibrillary tangle stage in a group with autopsy. We show that volume- and thickness-based measures generally perform similarly for separating clinically normal from AD populations, and in correlation with Braak neurofibrillary tangle stage at autopsy. Volume-based measures are generally more reliable than thickness measures. As expected, volume measures are highly correlated with head size, while thickness measures are generally not. Because approaches to statistically correcting volumes for head size vary and may be inadequate to deal with this underlying confound, and because our goal is to determine a measure which can be used to examine age and sex effects in a cohort across a large age range, we thus recommend thickness-based measures. Ultimately, based on these criteria and additional practical considerations of run-time and failure rates, we recommend an AD signature measure formed from a composite of thickness measurements in the entorhinal, fusiform, parahippocampal, mid-temporal, inferior-temporal, and angular gyrus ROIs using ANTs with input segmentations from SPM12.

Fig. 1

Performance of individual ROIs included in meta-ROIs for each of the five methods. For each method, regions in each row are ordered best to worst by their performance in the diagnostic separability (CN vs. MCI/AD) criteria, measured as an area under the receiver operating characteristic curve (AUROC), and plotted in the leftmost column. We show the six regions with the highest AUROC for each method, which together form the meta-ROIs. The second column plots the Spearman rank correlation (rho) between the ROI and total intracranial volume (TIV). The following two columns plot reliability of the measure in each ROI across successive scans, in the two Mayo and ADNI datasets respectively, as measured by intraclass correlation coefficients (ICC). In the ADNI reliability dataset, the ICCs are shown by manufacturer and indicated by colored points. The rightmost column plots the Spearman rank correlation between the ROI and negative Braak stage at autopsy. 95% confidence intervals are shown for all but the per-manufacturer Reliability-ADNI data. AUROC, rho, and ICC values are shown to the right of each panel. Numeric volumes listed for the ADNI reliability panel are for all manufacturers combined. Larger values (toward the right) are preferred in all columns with exception of the second, correlation with TIV, for which values near zero are preferred. Plots of performance for the top 20 ROIs by each method are included in the Supplementary material.

Fig. 2

Regions included in our recommended AD signature meta-ROIs using the SPM+DiReCT thickness method: entorhinal cortex, fusiform, parahippocampal, mid-temporal, inferior temporal, and angular gyrus.

Fig. 3

Comparative performance summary of AD Signature Meta-ROIs. A table of pair-wise p-values for comparisons between methods by each criterion is provided in Supplementary Table S2.

Fig. 4

Runtimes for candidate methods run on a single random Mayo subject scan.