Alzheimer's disease risk assessment using large-scale machine learning methods

Abstract

The goal of this work is to introduce new metrics to assess risk of Alzheimer's disease (AD) which we call AD Pattern Similarity (AD-PS) scores. These metrics are the conditional probabilities modeled by large-scale regularized logistic regression. The AD-PS scores derived from structural MRI and cognitive test data were tested across different situations using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. The scores were computed across groups of participants stratified by cognitive status, age and functional status. Cox proportional hazards regression was used to evaluate associations with the distribution of conversion times from mild cognitive impairment to AD. The performances of classifiers developed using data from different types of brain tissue were systematically characterized across cognitive status groups. We also explored the performance of anatomical and cognitive-anatomical composite scores generated by combining the outputs of classifiers developed using different types of data. In addition, we provide the AD-PS scores performance relative to other metrics used in the field including the Spatial Pattern of Abnormalities for Recognition of Early AD (SPARE-AD) index and total hippocampal volume for the variables examined.

Figure 1. The concept of a probabilistic hypercube is illustrated.

The probability hypercube can be interpreted as a geometrical representation of the output of a set of generative classifiers, each one estimated with different types of data. The set of AD-PS scores corresponding to a given individual define a position inside a unit hypercube. The position inside the hypercube for three individuals is illustrated.

Figure 2. RLR classifier performances across different types of information and cognitive groups.

Consistent with previous reports, grey matter (GM) tissue was more informative than white matter (WM) and cerebrospinal fluid (CSF). Interestingly, this difference decreases when a group with less severe cognitive decline is compared with the cognitively normal (CN) group.

Figure 3. The GM, WM and CSF discriminative maps produced by logistic regression with sparsity regularization are overlaid on the study customized template generated by DARTEL.

In each case, nine coronal slices (−82, −68, −52, −38, −22, −8, 8, 22, 38) are shown (neurological convention) in the first, second, and third rows, respectively. The blue areas are associated with AD classification, while the red ones are associated with CN classification.

Figure 4. Two-dimensional probabilistic hypercube views of ADNI data showing AD-PS grey matter (GM) and white matter (WM) scores for 188 cognitively normal (CN – blue stars) and 171 Alzheimer's disease (AD) patients (red circles).

They tended to cluster in different corners, as expected.