Biology subjects

Leenings, R.

Publications and source records attributed to Leenings, R..

4 recordsLinked to original sources

bioRxiv2026

Interpretable Hierarchical RNNs for rs-fMRI: Promise and Limits of Individualized Brain Dynamics

Modeling individual brain dynamics from resting-state fMRI (rs-fMRI) remains challenging due to substantial inter-subject variability, noise, and limited data length per subject. Here, we systematically evaluate whether hierarchical shallow piecewise-linear recurrent neural networks (shPLRNNs), recently introduced as interpretable dynamical system reconstruction models, can generate individualized rs-fMRI time series while preserving subject-specific functional connectivity structure. We applied the framework to 1,423 rs-fMRI samples from healthy participants of the Marburg-Munster Affective Disorders Cohort Study (MACS). Simulated rs-fMRI data reproduced substantial empirical FC structure, with comparable reconstruction accuracy on the validation and held-out test sets. Generalization to unseen individuals was heterogeneous and strongly depended on how typical a subjects connectivity pattern was relative to the training cohort, with template similarity explaining 37% of variance in reconstruction accuracy. Learned subject-specific parameters exhibited significant test-retest stability and higher within-subject than between-subject similarity on longitudinal data from two different timepoints, supporting their interpretation as individualized dynamical markers. Associations between individual parameters and demographic or cognitive variables were statistically significant but modest in effect size, and predictive performance remained below that obtained using empirical rs-fMRI features directly. Empirical FC was used as a reference for static subject information rather than as a target to be outperformed. Together, these results suggest that hierarchical shPLRNNs can extract meaningful and partially stable individual-specific dynamical structure from rs-fMRI data. The findings delineate key trade-offs between model expressivity, generalization and subject specificity, and point to directions for future methodological refinement in individualized brain modeling. Graphical AbstractA hierarchical dynamical RNN captures substantial individual rs-fMRI functional connectivity structure using compact subject-specific parameters embedded in shared population dynamics. The resulting representations generalize to held-out subjects and show test-retest stability, but only modest associations with phenotypic variables. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=150 SRC="FIGDIR/small/713153v3_ufig1.gif" ALT="Figure 1"> View larger version (61K): org.highwire.dtl.DTLVardef@1b029edorg.highwire.dtl.DTLVardef@90bdd6org.highwire.dtl.DTLVardef@9f6b77org.highwire.dtl.DTLVardef@486c9e_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗

bioRxiv2023

Cross-validation for the estimation of effect size generalizability in mass-univariate brain-wide association studies

IntroductionStatistical effect sizes are systematically overestimated in small samples, leading to poor generalizability and replicability of findings in all areas of research. Due to the large number of variables, this is particularly problematic in neuroimaging research. While cross-validation is frequently used in multivariate machine learning approaches to assess model generalizability and replicability, the benefits for mass-univariate brain analysis are yet unclear. We investigated the impact of cross-validation on effect size estimation in univariate voxel-based brain-wide associations, using body mass index (BMI) as an exemplary predictor. MethodsA total of n=3401 adults were pooled from three independent cohorts. Brain-wide associations between BMI and gray matter structure were tested using a standard linear mass-univariate voxel-based approach. First, a traditional non-cross-validated analysis was conducted to identify brain-wide effect sizes in the total sample (as an estimate of a realistic reference effect size). The impact of sample size (bootstrapped samples ranging from n=25 to n=3401) and cross-validation on effect size estimates was investigated across selected voxels with differing underlying effect sizes (including the brain-wide lowest effect size). Linear effects were estimated within training sets and then applied to unseen test set data, using 5-fold cross-validation. Resulting effect sizes (explained variance) were investigated. ResultsAnalysis in the total sample (n=3401) without cross-validation yielded mainly negative correlations between BMI and gray matter density with a maximum effect size of R2p=.036 (peak voxel in the cerebellum). Effects were overestimated exponentially with decreasing sample size, with effect sizes up to R2p=.535 in samples of n=25 for the voxel with the brain-wide largest effect and up to R2p=.429 for the voxel with the brain-wide smallest effect. When applying cross-validation, linear effects estimated in small samples did not generalize to an independent test set. For the largest brain-wide effect a minimum sample size of n=100 was required to start generalizing (explained variance >0 in unseen data), while n=400 were needed for smaller effects of R2p =.005 to generalize. For a voxel with an underlying null effect, linear effects found in non-cross-validated samples did not generalize to test sets even with the maximum sample size of n=3401. Effect size estimates obtained with and without cross-validation approached convergence in large samples. DiscussionCross-validation is a useful method to counteract the overestimation of effect size particularly in small samples and to assess the generalizability of effects. Train and test set effect sizes converge in large samples which likely reflects a good generalizability for models in such samples. While linear effects start generalizing to unseen data in samples of n>100 for large effect sizes, the generalization of smaller effects requires larger samples (n>400). Cross-validation should be applied in voxel-based mass-univariate analysis to foster accurate effect size estimation and improve replicability of neuroimaging findings. We provide open-source python code for this purpose (https://osf.io/cy7fp/?view_only=a10fd0ee7b914f50820b5265f65f0cdb).

neuroscience↗

bioRxiv2022

A Large-Scale ENIGMA Multisite Replication Study of Brain Age in Depression

BackgroundSeveral studies have evaluated whether depressed persons have older appearing brains than their nondepressed peers. However, the estimated neuroimaging-derived "brain age gap" has varied from study to study, likely driven by differences in training and testing sample (size), age range, and used modality/features. To validate our previously developed ENIGMA brain age model and the identified brain age gap, we aim to replicate the presence and effect size estimate previously found in the largest study in depression to date (N=2,126 controls & N=2,675 cases; +1.08 years [SE 0.22], Cohens d=0.14, 95% CI: 0.08-0.20), in independent cohorts that were not part of the original study. MethodsA previously trained brain age model (www.photon-ai.com/enigma_brainage) based on 77 FreeSurfer brain regions of interest was used to obtain unbiased brain age predictions in 751 controls and 766 persons with depression (18-75 years) from 13 new cohorts collected from 20 different scanners. ResultsOur ENIGMA MDD brain age model generalized reasonably well to controls from the new cohorts (predicted age vs. age: r = 0.73, R2=0.47, MAE=7.50 years), although the performance varied from cohort to cohort. In these new cohorts, on average, depressed persons showed a significantly higher brain age gap of +1 year (SE 0.35) (Cohens d{square}={square}{square}.15, 95% CI: 0.05-0.25) compared with controls, highly similar to our previous finding. ConclusionsThis study further validates our previously developed ENIGMA brain age algorithm. Importantly, we replicated the brain age gap in depression with a comparable effect size. Thus, two large-scale independent mega-analyses across in total 32 cohorts and >3,400 patients and >2,800 controls worldwide show reliable but subtle effects of brain aging in adult depression.

neuroscience↗

bioRxiv2019

Brain Aging in Major Depressive Disorder: Results from the ENIGMA Major Depressive Disorder working group

BackgroundMajor depressive disorder (MDD) is associated with an increased risk of brain atrophy, aging-related diseases, and mortality. We examined potential advanced brain aging in MDD patients, and whether this process is associated with clinical characteristics in a large multi-center international dataset. MethodsWe performed a mega-analysis by pooling brain measures derived from T1-weighted MRI scans from 29 samples worldwide. Normative brain aging was estimated by predicting chronological age (10-75 years) from 7 subcortical volumes, 34 cortical thickness and 34 surface area, lateral ventricles and total intracranial volume measures separately in 1,147 male and 1,386 female controls from the ENIGMA MDD working group. The learned model parameters were applied to 1,089 male controls and 1,167 depressed males, and 1,326 female controls and 2,044 depressed females to obtain independent unbiased brain-based age predictions. The difference between predicted "brain age" and chronological age was calculated to indicate brain predicted age difference (brain-PAD). FindingsOn average, MDD patients showed a higher brain-PAD of +0.90 (SE 0.21) years (Cohens d=0.12, 95% CI 0.06-0.17) compared to controls. Relative to controls, first-episode and currently depressed patients showed higher brain-PAD (+1.2 [0.3] years), and the largest effect was observed in those with late-onset depression (+1.7 [0.7] years). In addition, higher brain-PAD was associated with higher self-reported depressive symptomatology (b=0.05, p=0.004). InterpretationThis highly powered collaborative effort showed subtle patterns of abnormal structural brain aging in MDD. Substantial within-group variance and overlap between groups were observed. Longitudinal studies of MDD and somatic health outcomes are needed to further assess the predictive value of these brain-PAD estimates. FundingThis work was supported, in part, by NIH grants U54 EB020403 and R01 MH116147.

neuroscience↗