Age Related Differences in Cerebral Blood Flow and Cortical Thickness with an Application to Age Prediction

Abstract Cerebral cortex thinning and cerebral blood flow (CBF) reduction are typically observed during normal healthy aging. However, imaging-based age prediction models have primarily utilized morphological features of the brain. The addition of complementary physiological CBF information might therefore result in an improvement in age estimation. In this study, T1-weighted structural MRI and arterial spin labelling CBF images were acquired in 146 healthy participants across the adult lifespan. From each brain, 68 cerebral cortex regions were segmented and the cortical thickness and mean CBF were computed for each region. Linear regression with age was computed for each region and datatype and, laterality and correlation matrices were computed. Sixteen predictive models with-and-without feature selection were trained with the cortical thickness and CBF data alone as well as a combination of both data types. Age explained more variance in the cortical thickness data (average-R2 of 0.21) than in the CBF data (average-R2 of 0.09). When model calibration included feature selection, thirteen of sixteen model types performed slightly better with both datatypes. All 16 models performed significantly better when combining both measurement types and using feature selection, and thus, we conclude that the inclusion of CBF data marginally improves age estimation.