Predicting developmental norms from baseline cortical thickness in longitudinal studies.

Normative modeling has been applied to study how brain measures, such as gray matter thickness or volume, change across development. These models help identify how an individual's brain may differ from what is typical for their age or sex, which could eventually support more personalized treatments. However, most existing models use only one-time (cross-sectional) data, meaning they cannot capture how the brain changes over time. Longitudinal data, tracking the same individuals across multiple time points, is more informative but harder and more expensive to collect. We analyzed brain scans from over 6000 young people in the Adolescent Brain Cognitive Development (ABCD) study, about half of whom were girls. Each participant had brain scans at the start of the study, two and four years later. We deployed Baseline-Conditioned Norms (B-Norms) that used cortical thickness derived from each person's first scan and their ages at baseline and follow-up timepoint to predict cortical thickness at follow-up. We compared this to Cross-Sectional Norms (C-Norm), which only used age to predict thickness at follow-up. As expected, B-Norms predicted cortical thickness more accurately. Importantly, they were also better at detecting brain differences linked to puberty, especially in girls. Our findings suggest that our here proposed B-Norms may capture more developmental variance and may be more sensitive to sex-specific brain development over time during puberty. Therefore, B-norms may constitute a valuable complement to established C-norms.

backgroundNormative models have gained popularity in computational psychiatry for studying individual-level differences relative to population norms in biological data such as brain imaging, where measures like cortical thickness are typically predicted from variables such as age and sex. Nearly all published models to date are based on cross-sectional data, limiting their ability to predict longitudinal change.

methodsHere, we used longitudinal brain data from the Adolescent Brain Cognitive Development (ABCD) study, comprising cortical thickness measures from 180 regions per hemisphere in youths at baseline (N = 6179; 47% females), 2-year (N = 6179; 47% females), and 4-year (N = 805; 45% females) follow-up. A training set was established from baseline and 2-year follow-up data (N = 5374; 47% females), while data from individuals with all three time points available served as an independent test set (N = 805; 45% females). We developed sex-specific Baseline-Conditioned Norms (B-Norms) that predict brain region thickness at follow-up based on baseline thickness, baseline age, and follow-up age, and compared them to sex-specific Cross-Sectional Norms (C-Norms) that predict thickness at follow-up based on age alone.

resultsAs expected, out-of-sample testing in 2-year and 4-year follow-up data showed that B-Norms consistently provided better fits than C-Norms for nearly all cortical regions. Explained variance was higher in B-Norms than in C-Norms. No significant differences between time points (p = 0.45) were detected. Repeated measures ANOVA revealed differences in higher-order moments (e.g., skewness and kurtosis) for both models; for example, skewness varied by model, sex, time point, and their interactions. We showed that four regions were associated with pubertal changes in B-Norms but not in C-Norms, suggesting enhanced sensitivity of B-Norms to developmental processes.

conclusionTogether, our findings highlight the potential of B-Norms for capturing normative variation in longitudinal structural brain change, suggesting that they may constitute a valuable complement to existing C-Norms.