Emerging reports suggest that sample sizes commonly used in functional neuroimaging studies may be too small to detect many brain-behavior relationships, posing a major barrier to brain and mental health research. A central challenge is that planning robust studies requires researchers to know what effect sizes to expect, yet this essential information is surprisingly difficult to estimate in practice and thus often omitted from study planning. Critically, standard "mass univariate" procedures for estimating effects across multiple brain areas give an inflated picture of how large effect sizes are. Here, we introduce a method to correct this inflation bias and perform an unprecedented analysis of 63 studies in seven large datasets (n = 100-40,000; 52,979 total participants) to establish effect size benchmarks in functional neuroimaging. We find that between-subjects effects are exceedingly small at the majority of brain areas (Cohen's ∣d∣ < 0.2), requiring consortium-level sample sizes to detect even some of the strongest focal brain effects (n > 500 at 80% statistical power with FDR correction). However, multivariate analyses and within-subject task designs yield substantially larger effect sizes that can be detected at sample sizes within reach of individual labs (n < 50). By establishing data-driven effect size benchmarks, these findings lay the groundwork for more informed study planning in neuroscience while highlighting shared challenges (and the potential for shared solutions) across biomedicine.

Early life adversity (ELA) is a robust transdiagnostic risk factor for mental health disorders, yet the neurobiological mechanisms mediating its long-term impact remain poorly understood. Network control theory offers a novel framework for capturing how structural brain networks constrain and support brain dynamics. Controllability increases over development, associates with executive function and mental health, and appears sensitive to environmental insults. Thus, it may reflect a neurobiological mediator between ELA and behavioral outcomes. We tested whether alterations in modal controllability mediate the impact of multidimensional ELA on cognitive and behavioral outcomes in youth, and whether these pathways are shaped by genetic risk for neurodevelopmental conditions. Using data from 7,970 children aged 9-11 years in the Adolescent Brain Cognitive Development (ABCD) Study, we derived five latent ELA dimensions from 67 indicators, and computed polygenic risk scores (PRS) for attention-deficit/hyperactivity disorder (ADHDPRS) and autism spectrum disorder (ASDPRS). Distinct ELA dimensions were associated with increased controllability in medial frontal, frontoparietal, default mode, and motor networks, as well as with externalizing symptoms and impaired crystallized cognition. Controllability partially mediated these associations, and indirect effects were significantly moderated ADHDPRS and ASDPRS. Longitudinal analyses further demonstrated that baseline controllability predicted cognitive performance two years later. These findings delineate a neurodevelopmental cascade linking early adversity and genetic vulnerability to transdiagnostic mental health risk, positioning brain controllability as a promising mechanistic marker and potential target for early intervention.

Prevention remains a key strategy to address the growing burden of metabolic dysfunction-associated steatotic liver disease (MASLD), highlighting the importance of exploring modifiable risk factors. Accumulating evidence suggests a close link between physical frailty and MASLD. However, how frailty interacts with metabolic syndrome to affect MASLD and the causality and direction of the association remain largely unknown. Leveraging data from 405,224 UK Biobank participants with a 13.65-year follow-up, we found that physical frailty was associated with an increased risk of clinically diagnosed MASLD and exacerbated the adverse effect of metabolic syndrome on MASLD incidence, implying that frail people may be more vulnerable to this disease because of metabolic syndrome. Mendelian randomization provided evidence for a potential causal effect of physical frailty on MASLD but not the reverse direction. Moreover, the metabolome-wide association analysis revealed widespread associations of plasma metabolites with both frailty and MASLD, suggesting a shared metabolomic foundation between them. Some metabolites, including fatty acids and triglyceride-rich lipoprotein biomarkers, partially explained the frailty-MASLD relationship, indicating a potential metabolomic mechanism. If confirmed in further studies, frailty screening may help identify high-risk individuals and inform early prevention for MASLD, especially for those with metabolic syndrome.

Reading ability depends on multiple cognitive skills, including decoding and language comprehension, which can vary widely across individuals-even among those with similarly low reading performance. To better understand the brain basis of this variability, we used connectome-based predictive modeling (CPM) to identify large-scale functional connectivity patterns associated with reading and language skills in a population-based sample. Cross-sectional CPM models were trained using functional connectivity data from the Adolescent Brain and Cognitive Development study (n = 6894) and tested in two independent cohorts: the New Haven Lexinome Project and the Genes, Reading, and Dyslexia study (combined n = 136). Functional connectivity measures included both resting- and task-based scans. Reading and language were measured with psychometric tests of word reading and vocabulary, respectively. CPM models significantly predicted reading (r = .24) and language (r = .28) scores in the discovery sample and generalized to an external sample (rs = .23 and .19). Anatomically, the reading and language models showed significant overlap, with the medial frontal network emerging as most predictive in both. However, these models exhibited distinct generalization patterns to children with decoding versus language comprehension difficulties-classified using 20th percentile cutoffs-highlighting their neural specificity. Reading and language models included distinct connectivity signatures and generalized differently to children with decoding versus language comprehension difficulties. These findings demonstrate that although reading and language abilities are behaviorally related, they are supported by partially distinct neural architectures. Integrating behavioral and neuroimaging data may clarify specific brain-behavior relationships and inform more tailored interventions for children with reading and language difficulties.

Chronic pain conditions frequently coexist and share common genetic vulnerabilities. Despite evidence showing associations between pain and depression, the additive effect of co-occurring pain conditions on depression risk and the underlying mechanisms remain unclear. Leveraging data from 431,038 UK Biobank participants with 14-year follow-up, we found a significantly increased risk of depression incidence in individuals reporting pain, irrespective of body site or duration (acute or chronic), compared with pain-free individuals. The depression risk increased with the number of co-occurring pain sites. Mendelian randomization supported potential causal inference. We constructed a composite pain score by combining individual effects of acute or chronic pain conditions across eight body sites in a weighted manner. We found that depression risks increased monotonically in parallel with composite pain scores. Moreover, some inflammatory markers, including C-reactive protein, partially mediated the association between composite pain scores and depression risk. Considering the high prevalence of comorbid depression and pain, pain screening may help identify high-risk individuals for depression.

High-amplitude coactivation patterns are sparsely present during resting-state functional magnetic resonance imaging (fMRI), yet they drive functional connectivity and resemble task activation patterns. However, little research has characterized the remaining majority of the resting-state signal. Here, we introduce caricaturing, a method for projecting resting-state data onto a subspace orthogonal to a manifold of coactivation patterns estimated from task fMRI data. This removes linear combinations of these coactivation patterns from resting-state data to create caricatured connectomes. We used task data from two large-scale neuroimaging datasets to construct a manifold of task coactivation patterns and created caricatured connectomes. These connectomes exhibit lower between-individual similarity and higher identifiability and could be used to predict phenotypic measures, representing individual differences in behavior, often to a greater degree than standard connectomes. Our results show that there is a useful signal beyond the dominant coactivations that drive resting-state functional connectivity, which may better characterize the brain's intrinsic functional architecture.

While the world is aware of America's history of enslavement, the ongoing impact of anti-Black racism in the United States remains underemphasized in health intervention modeling. This Perspective argues that algorithmic bias—manifested in the worsened performance of clinical algorithms for Black vs. white patients—is significantly driven by the failure to model the cumulative impacts of racism-related stress, particularly racial heteroscedasticity. Racial heteroscedasticity refers to the unequal variance in health outcomes and algorithmic predictions across racial groups, driven by differential exposure to racism-related stress. This may be particularly salient for Black Americans, where anti-Black bias has wide-ranging impacts that interact with differing backgrounds of generational trauma, socioeconomic status, and other social factors, promoting unaccounted for sources of variance that are not easily captured with a blanket “race” factor. Not accounting for these factors deteriorates performance for these clinical algorithms for all Black patients. We outline key principles for anti-racist AI governance in healthcare, including: (1) mandating the inclusion of Black researchers and community members in AI development; (2) implementing rigorous audits to assess anti-Black bias; (3) requiring transparency in how algorithms process race-related data; and (4) establishing accountability measures that prioritize equitable outcomes for Black patients. By integrating these principles, AI can be developed to produce more equitable and culturally responsive healthcare interventions. This anti-racist approach challenges policymakers, researchers, clinicians, and AI developers to fundamentally rethink how AI is created, used, and regulated in healthcare, with profound implications for health policy, clinical practice, and patient outcomes across all medical domains.

Objective:Conduct a mega-analysis of two complementary measures of resting-state functional magnetic resonance imaging (rsfMRI) dynamics-amplitude of low-frequency fluctuation (ALFF) and low-frequency spectral entropy (lfSE)-in mood and psychosis-spectrum disorders to evaluate group differences and clinical symptom associations. Design:ALFF and lfSE were calculated at the node-level by filtering data from 0.01 Hz to 0.08 Hz, regressing demographic variables, and harmonizing sites. Group differences were assessed using the Wilcoxon signed test. Symptom associations were evaluated with Spearman's rho. Analyses were conducted at both whole-brain and network levels, with sensitivity analyses to evaluate the impact of frequency brands. Setting:Four independent open-source case-control datasets with resting-state functional magnetic resonance imaging were used: the Center for Biomedical Research Excellence, the Human Connectome Project for Early Psychosis, the Strategic Research Program for Brain Sciences, and the UCLA Consortium for Neuropsychiatric Phenomics. Participants:Included participants had a mood disorder (bipolar, dysthymia, or major depressive disorder, n=228, aged 38.31 ± 12.56 years), a psychosis-spectrum disorder (early psychosis, schizophrenia spectrum disorder, or mood disorder with psychotic symptoms, n=318, aged 29.8 ± 13.21 years), or a healthy control (n=535, aged 39.89 ± 15.3 years). Main outcomes and Measures:To identify group differences and symptom associations in mood and psychosis-spectrum disorders using ALFF and lfSE. Results:ALFF in psychosis-spectrum was significantly lower than mood disorders and controls (q's<0.001) at the whole-brain and network levels. lfSE in controls was significantly lower than both psychosis-spectrum and mood disorders at the whole-brain and network levels (q's<0.001). Whole-brain ALFF is positively associated with mood symptoms (rho=0.27, p<0.05). Whole-brain lfSE is negatively associated with positive (rho=-0.13, p<0.05) and mood (rho=-0.38, p<0.01) symptoms. A greater sensitivity of group differences and symptom associations to frequency ranges was observed in mood disorders. ALFF is sensitive to medication. Conclusions and Relevance:Widespread, global differences in ALFF and lfSE underly psychosis-spectrum and mood disorders. lfSE may be applicable for wider use in fMRI. Differences in spectral measures of brain dynamics may represent shared and distinct markers of mental health.

BACKGROUND:Physical frailty is a state of increased vulnerability to stressors and is associated with serious health issues. However, how frailty affects and is affected by numerous other factors, including mental health and brain structure, remains underexplored. We aimed to investigate the mutual effects of frailty and health using large, multidimensional data.METHODS:For this population-based study, we used data from the UK Biobank to examine the pattern and direction of association between physical frailty and 325 health-related measures across multiple domains, using linear mixed-effect models and adjusting for numerous confounders. Participants were included if complete data were available for all five indicators of frailty, all covariates, and at least one health measure. We further examined the association between frailty and brain structure and the role of this association in mediating the relationship between frailty and health outcomes.FINDINGS:483 033 participants aged 38-73 years were included in the study at baseline (between Dec 19, 2006, and Oct 1, 2010); at a median follow-up of 9 years (IQR 8-10), behavioural data were available for 46 501 participants and neuroimaging data for 40 210 participants. The severity of physical frailty was significantly associated with decreased cognitive performance (Cohen's d=0·025-0·162), increased early-life risks (d=0·026-0·111), unhealthy lifestyle (d=0·013-0·394), poor physical fitness (d=0·007-0·668), increased symptoms of poor mental health (d=0·032-0·607), severe environmental pollution (d=0·013-0·064), and adverse biochemical markers (d=0·025-0·198). Some associations were bidirectional, with the strongest effects on mental health measures. The severity of frailty correlated with increased total white matter hyperintensity and lower grey matter volume, particularly in subcortical regions (d=0·027-0·082), which significantly mediated the association between frailty and health-related outcomes, although the mediated effects were small.INTERPRETATION:Physical frailty is associated with diverse unfavourable health-related outcomes, which can be mediated by differences in brain structure. Our findings offer a framework for guiding preventative strategies targeting both frailty and psychiatric disorders.FUNDING:National Institute of Mental Health, National Science Foundation.

The prevalence of machine learning in biomedical research is rapidly growing, yet the trustworthiness of such research is often overlooked. While some previous works have investigated the ability of adversarial attacks to degrade model performance in medical imaging, the ability to falsely improve performance via recently-developed "enhancement attacks" may be a greater threat to biomedical machine learning. In the spirit of developing attacks to better understand trustworthiness, we developed two techniques to drastically enhance prediction performance of classifiers with minimal changes to features: 1) general enhancement of prediction performance, and 2) enhancement of a particular method over another. Our enhancement framework falsely improved classifiers' accuracy from 50% to almost 100% while maintaining high feature similarities between original and enhanced data (Pearson's r's > 0.99). Similarly, the method-specific enhancement framework was effective in falsely improving the performance of one method over another. For example, a simple neural network outperformed logistic regression by 17% on our enhanced dataset, although no performance differences were present in the original dataset. Crucially, the original and enhanced data were still similar (r = 0.99). Our results demonstrate the feasibility of minor data manipulations to achieve any desired prediction performance, which presents an interesting ethical challenge for the future of biomedical machine learning. These findings emphasize the need for more robust data provenance tracking and other precautionary measures to ensure the integrity of biomedical machine learning research. Code is available at https://github.com/mattrosenblatt7/enhancement_EPIMI.

Connectome-based predictive models are widely used in the neuroimaging community and hold great clinical potential. Recent literature has focused on improving the accuracy and fairness of connectome-based models, while largely overlooking trustworthiness, defined as the robustness of a model to data manipulations. In this work, we investigate the idea of trustworthiness through backdoor data poisoning-a technique that manipulates a portion of the training data to encourage misclassification of a specific subset of testing data, while all other testing data remain unaffected. Furthermore, we demonstrate two defenses that mitigate, but do not completely prevent, the effects of data poisoning: randomized discretization and leave-one-site-out ensemble detection. Our findings suggest that trustworthiness in connectome-based predictive models needs to be carefully evaluated before any clinical applications and that defenses are necessary to ensure model outputs are trustworthy. Code is available at https://github.com/mattrosenblatt7/connectome_poisoning.