Pitfalls in brain age analyses

Abstract

Over the past decade, there has been an abundance of research on the difference between age and age predicted using brain features, which is commonly referred to as the “brain age gap.” Researchers have identified that the brain age gap, as a linear transformation of an out‐of‐sample residual, is dependent on age. As such, any group differences on the brain age gap could simply be due to group differences on age. To mitigate the brain age gap's dependence on age, it has been proposed that age be regressed out of the brain age gap. If this modified brain age gap is treated as a corrected deviation from age, model accuracy statistics such as R2 will be artificially inflated to the extent that it is highly improbable that an R2 value below .85 will be obtained no matter the true model accuracy. Given the limitations of proposed brain age analyses, further theoretical work is warranted to determine the best way to quantify deviation from normality.

Detection of advanced brain aging in schizophrenia and its structural underpinning by using normative brain age metrics

“…Although some correction methods have been proposed to remove the bias based on a regression adjustment of the PAD ( Beheshti et al, 2019 , de Lange and Cole, 2020 ), these methods are prone to artificially inflate the model accuracy and have inherent circularity of age and age prediction, leading to over- or underestimated results. ( Butler et al, 2021 ). The framework of nPAD might offer a solution to these limitations; nPAD is theoretically free of age-related bias in terms of its definition because the reference of comparison of one’s brain age is the peers’ brain age.…”

Deep Relation Learning for Regression and Its Application to Brain Age Estimation

“…This bias, or RTM problem, also exists in other age estimation studies that focus purely on adults, where bias correction has been proposed [16], [14], [17]. Bias correction, however, is also with controversies, for which model should be optimal for correction, how to best quantify bias, and how to best evaluate the effects of correction [18]. Bias correction is also our ongoing work.…”

“…Based on the regression to the mean problem [15], the brain ages of the young subjects are usually over-estimated and the brain ages of the old subjects are usually under-estimated. Recent studies are on correcting this bias [16], [14], [17], but with controversies [18]. With the pair of the young and old subjects, the cumulative relation we propose here may serve as a potential model for bias correction.…”

Factors Influencing Change in Brain-Predicted Age Difference in a Cohort of Healthy Older Individuals

“…While not a statistically significant, the weak negative correlation between brain-PAD and chronological age may point to an age bias (i.e., statistical phenomenon whereby the model underestimates brain age in older populations), that could partially explain the negative mean brain-PAD observed for our study cohort and may confound findings with other age-related exposures [ 37 ]. In attempt to overcome this limitation, we include chronological age as a covariate in all mixed models [ 96 ]. Finally, though estimates of brain age were derived from local tissue volume (i.e., grey and white matter volumes) from across the whole brain, which change with aging [ 3 ], they can provide no evidence on associations with the change in regional brain aging.…”