Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum: Development of the Classifier and Longitudinal Evaluation

Abstract

To develop a new method for measuring Alzheimer's disease (AD)-specific similarity of cortical atrophy patterns at the individual-level, we employed an individual-level machine learning algorithm. A total of 869 cognitively normal (CN) individuals and 473 patients with probable AD dementia who underwent high-resolution 3T brain MRI were included. We propose a machine learning-based method for measuring the similarity of an individual subject's cortical atrophy pattern with that of a representative AD patient cohort. In addition, we validated this similarity measure in two longitudinal cohorts consisting of 79 patients with amnestic-mild cognitive impairment (aMCI) and 27 patients with probable AD dementia. Surface-based morphometry classifier for discriminating AD from CN showed sensitivity and specificity values of 87.1% and 93.3%, respectively. In the longitudinal validation study, aMCI-converts had higher atrophy similarity at both baseline (p < 0.001) and first year visits (p < 0.001) relative to non-converters. Similarly, AD patients with faster decline had higher atrophy similarity than slower decliners at baseline (p = 0.042), first year (p = 0.028), and third year visits (p = 0.027). The AD-specific atrophy similarity measure is a novel approach for the prediction of dementia risk and for the evaluation of AD trajectories on an individual subject level.

Figure 1

Discriminating features of our classification. (A) The discriminating regions of our classification on the atlas surface meshes and (B) The discriminative pattern of each patient with aMCI and AD. Color intensities in the figure represent discriminative power in AD classification. aMCI = amnestic mild cognitive impairment; AD = Alzheimer’s disease.

Figure 2

Comparisons of the AD-specific atrophy similarity at baseline and follow-up years: (A) non-converters vs. converters in patients with aMCI and (B) slow- and fast-decliners in patients with AD. Mixed effects models of the worsening in AD-specific atrophy similarity over time between the classified groups by clinical progression in patients with aMCI and AD showed significant differences between the groups (p = 0.027 in aMCI cohort and p = 0.029 in AD cohort). aMCI = amnestic mild cognitive impairment; AD = Alzheimer’s disease.

Figure 3

Overview of the proposed method. (A) Image preprocessing; (B) Group classifier training; and (C) AD-specific pattern similarity computation. AD = Alzheimer’s disease.

Figure 4

Examples of AD-specific atrophy similarity measure at the individual-level. The AD-specific atrophy similarity scores differed between Case #96 - CN (left, 3.7) and Case #1256 - AD (right, 91.6). The standardized value (Z-score) maps were computed to visualize the AD-specific atrophy similarity. Positive Z-scores (red) indicate that the regions of brain are similar to the AD-specific patterns of atrophy. AD = Alzheimer’s disease; CN = cognitively normal; MMSE = mini-mental state examination.