Mapping abnormal subcortical brain morphometry in an elderly HIV+ cohort

Abstract

Over 50% of HIV + individuals exhibit neurocognitive impairment and subcortical atrophy, but the profile of brain abnormalities associated with HIV is still poorly understood. Using surface-based shape analyses, we mapped the 3D profile of subcortical morphometry in 63 elderly HIV + participants and 31 uninfected controls. The thalamus, caudate, putamen, pallidum, hippocampus, amygdala, brainstem, accumbens, callosum and ventricles were segmented from high-resolution MRIs. To investigate shape-based morphometry, we analyzed the Jacobian determinant (JD) and radial distances (RD) defined on each region's surfaces. We also investigated effects of nadir CD4 + T-cell counts, viral load, time since diagnosis (TSD) and cognition on subcortical morphology. Lastly, we explored whether HIV + participants were distinguishable from unaffected controls in a machine learning context. All shape and volume features were included in a random forest (RF) model. The model was validated with 2-fold cross-validation. Volumes of HIV + participants' bilateral thalamus, left pallidum, left putamen and callosum were significantly reduced while ventricular spaces were enlarged. Significant shape variation was associated with HIV status, TSD and the Wechsler adult intelligence scale. HIV + people had diffuse atrophy, particularly in the caudate, putamen, hippocampus and thalamus. Unexpectedly, extended TSD was associated with increased thickness of the anterior right pallidum. In the classification of HIV + participants vs. controls, our RF model attained an area under the curve of 72%.

Keywords: CD4, cluster of differentiation; Classification; HIV; JD, Jacobian determinant; MRI; RD, radial distance; Random forest; Shape analysis; Subcortical.

Fig. 1

Scatterplot of volumetric differences between HIV + and HIV − participants. The plots are divided into rows showing structures in the left hemisphere (top), midline (middle) and right hemisphere (bottom). All plots have been centered and scaled. The HIV status coefficient (in mm³) resulting from Eq. (1) is given for each plot. Volumetric differences that survive FDR correction are highlighted in red.

Fig. 2

FDR-thresholded t-value maps of the HIV + vs. HIV − contrast. (a) maps the significant differences in RD while (b) maps significant differences in JD. Images are shown in radiological orientation (i.e. left-right flipped), orientations are provided beside each set of surfaces.

Fig. 3

FDR-thresholded t-value maps of the association between TSD and (a) RD and (b) JD. Images are in radiological orientation (i.e. left–right flipped). The first column shows an anterior view while the second is the medial view of the right hemisphere structures.

Fig. 4

FDR-thresholded t-value maps of (a–b) association of thickness and the interaction of WAIS score and HIV Status, (c–d) the main effect of WAIS on thickness and (e–f) the association of WAIS and thickness within the HIV + cohort only.

Fig. 5

ROC curves and associated AUC values for random forest classification using the full set of morphological features. ROCs for feature subsets are also provided.

Fig. 6

Mapping of random forest importance scores in the classification of HIV status to subcortical surfaces. As all features were included in the same RF model, their importance is with respect to all other features. (a) RD and (b) JD feature set importance scores. Images are in radiological orientation (i.e. left–right flipped), orientations are provided beside each set of surfaces.

Fig. 7

Average random forest importance scores by region and feature set.