25 January 2024 in Journal of Systems Science and Complexity
1 December 2025 in bioRxiv
Machine learning is used in neuroscience to examine brain-phenotype associations and facilitate individual prediction from high-dimensional brain imaging. For continuous phenotypes, Pearson correlation between the observed and predicted phenotype is used to quantify model accuracy in testing data. However, recent research suggests millions of samples may be needed to reliably estimate the maximum achievable predictive accuracy (MAPA). We formally define the MAPA and show that the Pearson estimator is biased for this quantity and its confidence intervals fail to capture the target. We develop a semiparametric (double machine learning) one-step estimator that more accurately estimates the MAPA and yields valid confidence intervals across flexible machine learning settings. Analyzing data from the Reproducible Brain Charts dataset, we show that this estimator has smaller bias when estimating brain-phenotype associations of neuroimaging data with age and psychopathology phenotypes. We show that MAPA for psychopathology factor scores using machine learning models built on structural and functional imaging measures is not better than using demographic and nuisance covariates alone.
23 January 2026 in Research in Autism
Autistic youth frequently experience co-occurring anxiety. In non-autistic populations, females are more likely to be diagnosed with anxiety and demonstrate distinct symptom trajectories across adolescence. Prior research has demonstrated diagnostic group differences in parent- and self-reported anxiety symptoms; however, it is largely unknown if and how symptoms change over puberty. We examine trajectories of anxiety symptoms for autistic and non-autistic youth over a four-year (Y1-Y4) longitudinal study. Autistic (N=140) and non-autistic (N=104) youth (ages 10-13 years at Y1) completed the parent- and self-report versions of the Multidimensional Anxiety Scale for Children, 2nd Edition (MASC-2). Linear mixed effects models were used to examine the impact of age, sex, and diagnosis on anxiety symptoms. There was a significant diagnosis-sex interaction (p FDR =0.005) on the parent-report of anxiety and a main effect of age (p<0.001). Autistic and non-autistic females showed an increase in anxiety symptoms over time; autistic males showed a slight decrease. There was a main effect of diagnosis (p FDR =0.003) on the child-report measure of anxiety symptoms; the effects of age (p FDR =0.117) and sex were not significant (p FDR =0.305). The relationship between pubertal development and anxiety symptoms differed between the subgroups. Our main findings - that autistic youth demonstrate greater anxiety symptoms than non-autistic youth, with distinct symptom trajectories emerging in females versus males - add to the growing literature aimed at understanding the biopsychosocial factors impacting puberty and emotional functioning across adolescence.
11 February 2025 in Statistical Methods in Medical Research
We conducted a systematic comparison of statistical methods used for the analysis of time-to-event outcomes under various proportional and non-proportional hazard (NPH) scenarios. Our study used data from recently published oncology trials to compare the Log-rank test, still by far the most widely used option, against some available alternatives, including the MaxCombo test, the Restricted Mean Survival Time difference test, the Generalized Gamma model and the Generalized F model. Power, type I error rate, and time-dependent bias with respect to the survival probability and median survival time were used to evaluate and compare the performance of these methods. In addition to the real data, we simulated three hypothetical scenarios with crossing hazards chosen so that the early and late effects “cancel out” and used them to evaluate the ability of the aforementioned methods to detect time-specific and overall treatment effects. We implemented novel metrics for assessing the time-dependent bias in treatment effect estimates to provide a more comprehensive evaluation in NPH scenarios. Recommendations under each NPH scenario are provided by examining the type I error rate, power, and time-dependent bias associated with each statistical approach.
12 February 2025 in bioRxiv
The majority of neuroimaging inference focuses on hypothesis testing rather than effect estimation. With concerns about replicability, there is growing interest in reporting standardized effect sizes from neuroimaging group-level analyses. Confidence sets are a recently developed approach to perform inference for effect sizes in neuroimaging but are restricted to univariate effect sizes and cross-sectional data. Thus, existing methods exclude increasingly common multigroup or nonlinear longitudinal associations of biological brain measurements with inter- and intra-individual variations in diagnosis, development, or symptoms. We broadly generalize the confidence set approach by developing a method for arbitrary effect sizes in longitudinal studies. Our method involves robust estimation of the effect size image and spatial and temporal covariance function based on generalized estimating equations. We obtain more efficient effect size estimates by concurrently estimating the exchangeable working covariance and using a nonparametric bootstrap to determine the joint distribution of effect size across voxels used to construct confidence sets. These confidence sets identify regions of the image where the lower or upper simultaneous confidence interval is above or below a given threshold with high probability. We evaluate the coverage and simultaneous confidence interval width of the proposed procedures using realistic simulations and perform longitudinal analyses of aging and diagnostic differences of cortical thickness in Alzheimer’s disease and diagnostic differences of resting-state hippocampal activity in psychosis. This comprehensive approach along with the visualization functions integrated into the pbj R package offers a robust tool for analyzing repeated neuroimaging measurements.