---
title: "Predicted Brain Age Differences Across Nine Disorders: Case-Control MRI Study"
id: "plos-medicine-1-brain-aging-patterns-among-nine-neurological-disorders-a-case-control-study"
canonical_url: "https://medichelpline.com/clinical-feed/plos-medicine-1-brain-aging-patterns-among-nine-neurological-disorders-a-case-control-study"
content_type: "clinical_feed_article"
specialty: "Neurology"
source_name: "PLOS Medicine"
source_url: "https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1004860"
published_at: "2026-07-21T14:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Predicted Brain Age Differences Across Nine Disorders: Case-Control MRI Study
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/plos-medicine-1-brain-aging-patterns-among-nine-neurological-disorders-a-case-control-study
- **Specialty:** [Neurology](https://medichelpline.com/clinical-feed/neurology.md)
- **Primary Source:** PLOS Medicine
- **Source URL:** [Original Journal Publication](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1004860)
- **Published At:** 2026-07-21T14:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- This case-control study used structural MRI from 45,900 healthy controls and 2,698 patients spanning nine common brain disorders to compare neuroimaging-predicted brain age with chronological age, using the **predicted age difference (PAD)** as the main biomarker. - Diagnostic groups included developmental disorders (ADHD, ASD), addiction (alcohol use disorder [AUD], tobacco use disorder [TUD], and combined AUD&TUD), dementia (Alzheimer’s disease [AD], mild cognitive impairment [MCI]), and psychiatric disorders (schizophrenia [SZ], bipolar disorder [BP], major depressive disorder [MDD]). - PAD was derived from XGBoost models trained on age-matched healthy samples and adjusted in linear models for age, age2, sex, and site; group effects were summarized as Cohen’s d with 95% confidence intervals. - Dementia showed the largest accentuated aging: AD d = 0.97 (95% CI 0.82–1.13), MCI d = 0.45 (95% CI 0.34–0.56); addiction groups had high PAD (combined A&TUD d = 0.84, TUD d = 0.72, AUD d = 0.62), and psychiatric disorders showed moderate increases (SZ d = 0.53, BP d = 0.46, MDD d = 0.28). - Developmental disorders did not show significant PAD differences (ASD d = 0.06, p = 0.36; ADHD d = 0.01, p = 0.98). - Disorder-specific spatial patterns associated with higher PAD included the **prefrontal cortex** across disorders; psychiatric disorders linked to frontotemporal circuits, addiction to default-mode/salience-putamen-thalamus networks, and dementia to fronto‑occipital networks. - Transcriptome enrichment analyses showed that genes associated with each disorder-specific PAD pattern were enriched in distinct biological processes; exact gene lists and enrichment results were reported in the source but not restated here. - The study notes a key limitation: high comorbidity between psychiatric disorders and addiction was not modeled and may confound findings. - Data provenance: training data from HCP, GSP, UKB; testing data included public consortia (ADNI, ABIDE, ADHD-200) and protected datasets for psychiatric/addiction cohorts; code and de-identified minimal data are available on GitHub and Zenodo per source.
## Clinical Analysis & Structured Key Points
SKIP TO MAIN CONTENT Advertisement plos.org Create account Sign in About Browse Publish advanced search 0 Save 0 Citation 61 View 0 Share OPEN ACCESS PEER-REVIEWED RESEARCH ARTICLE Brain aging patterns among nine neurological disorders: A case-control study Chuang Liang , Godfrey Pearlson , Juan Bustillo, Peter Kochunov, Jiayu Chen, Xiangrong Zhang, Rongtao Jiang, Kent E. Hutchison, Jing Sui, Zening Fu, Xiao Yang, Yuhui Du, Daoqiang Zhang, Shile Qi , Vince D. Calhoun Published: July 21, 2026 https://doi.org/10.1371/journal.pmed.1004860 Article Authors Metrics Comments Media Coverage Abstract Author summary Introduction Methods Results Discussion Supporting information Acknowledgments References Reader Comments Figures Abstract Background The difference between neuroimaging-predicted brain age and chronological age, the predicted age difference (PAD), has been studied as a potential biomarker reflecting individual brain health. Although previous large-scale studies have shown that brain age deviations occur across multiple disorders, cross-disorder comparisons of PAD within a unified framework, together with identification of the neuroimaging features associated with these differences and their related gene expression profiles, remain limited. Our aims are to systematically compare brain aging across multiple common brain disorders and explore the brain patterns and biological processes underlying these differences. Methods and findings In this study, structural MRI data from 45,900 healthy controls (HCs) and 2,698 patients with developmental disorders (attention-deficit/hyperactivity disorder [ADHD] and autism spectrum disorder [ASD]), addiction (alcohol use disorder [AUD], tobacco use disorder [TUD], and AUD&TUD-A&TUD), dementia (Alzheimer’s disease [AD], and mild cognitive impairment [MCI]) or other psychiatric disorders (schizophrenia [SZ], bipolar disorder [BP], and major depressive disorder [MDD]), were collected to generate PAD, along with transcriptome data. Then, we calculated the PAD difference between patient and HC as Cohen’s d effect sizes, derived from a linear model that accounted for age, age2, sex, and site, and further identified the interpretable brain patterns associated with the PAD difference for each diagnostic group. Finally, enrichment analyses was conducted to identify the biological function of genes relatively over- or underexpressed in association with these patterns. Results showed that while PAD was consistently greater across disorders, different brain disorders showed different degrees of abnormality, the highest effects in dementia (AD: d = 0.97, 95% confidence interval (CI) [0.82,1.13]; p < 0.001 and MCI: d = 0.45, 95% CI [0.34,0.56]; p < 0.001), followed by addiction (A&TUD: d = 0.84, 95% CI [0.44,1.23]; p < 0.001, TUD: d = 0.72, 95% CI [0.49,0.96]; p < 0.001, and AUD d = 0.62, 95% CI [0.39,0.84]; p < 0.001) and psychiatric disorders (SZ: d = 0.53, 95% CI [0.30,0.76]; p < 0.001, BP: d = 0.46, 95% CI [0.22,0.69]; p < 0.001 and MDD: d = 0.28, 95% CI [0.11,0.46]; p < 0.001), but not different from expected in developmental disorders (ASD: d = 0.06, 95% CI [−0.04,0.16]; p = 0.36) and ADHD: d = 0.01, 95% CI [−0.14,0.15]; p = 0.98). Furthermore, higher PAD values in patient groups were linked to specific spatial brain patterns, including the frontotemporal network in psychiatric disorders, default mode network-salience network-putamen-thalamus in addiction and fronto-occipital network in dementia. Prefrontal cortex involvement was common across disorders, and disorder-specific brain patterns associated genes were enriched in different biological processes. A limitation of our study is that psychiatric disorders and addiction have high comorbidity, and these potential confounders were not considered. Conclusions In summary, the different brain aging patterns, each based around specific underlying circuits, may serve as neuroimaging biomarkers for understanding the neural aging mechanisms in commonly occurring brain disorders. Future studies should test whether these disorder-specific brain aging patterns can serve as useful biomarkers to guide critical clinical decision-making. Author summary Why was this study done? The difference between brain age estimated from MRI scans and actual age (PAD) can reflect brain health. Previous studies have shown that many common brain disorders are linked to accentuated brain aging (increased PAD compared to healthy control (HC)). However, it remains unclear whether brain aging follows similar or distinct patterns across different disorders and how these differences can be interpreted biologically. We aimed to systematically compare brain aging across multiple common brain disorders and explore the brain patterns and biological processes underlying these differences. What did the researchers do and find? We analyzed brain imaging data from 45,900 healthy individuals and 2,698 patients with developmental disorders, addiction, dementia, or other psychiatric disorders, to compare the PAD difference between patient and HC and identify the brain regions and biological processes underlying these differences. We found that brain age was generally higher in patients, with the largest increases observed in dementia, followed by addiction and psychiatric disorders, but not in developmental disorders. Different disorders showed distinct brain regions associated with these differences, with a shared involvement of the prefrontal cortex, and these related genes were linked to different biological functions. What do these findings mean? Accentuated brain aging is a common result for many disorders, with the prefrontal cortex may underlie the important neural processes with respect to the universality of accentuated brain aging. The biological processes of genes related to identified brain patterns can help us to better understand brain alterations related to aging at a molecular level. In the future, the brain patterns we identified for each disorder could be tested for their usefulness as biomarkers to guide critical clinical decision-making. The main limitation is that we did not account for comorbidities between psychiatric disorders and addiction, which may have introduced potential confounding effects and influenced the observed results. Figures Citation: Liang C, Pearlson G, Bustillo J, Kochunov P, Chen J, Zhang X, et al. (2026) Brain aging patterns among nine neurological disorders: A case-control study. PLoS Med 23(7): e1004860. https://doi.org/10.1371/journal.pmed.1004860 Academic Editor: Carol Brayne, University of Cambridge, UNITED KINGDOM OF GREAT BRITAIN AND NORTHERN IRELAND Received: November 28, 2025; Accepted: June 15, 2026; Published: July 21, 2026 Copyright: © 2026 Liang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The main code used in this study is available at https://github.com/liangchuang11/Brain-age-prediction.git and has been archived with a citable DOI on Zenodo: https://doi.org/10.5281/zenodo.20743298. The de-identified minimal data required to replicate the findings of this study have been made publicly available without restriction on the GitHub repository: https://github.com/liangchuang11/brain_age_de-identified-minimal_data.git and archived with a DOI on Zenodo: https://doi.org/10.5281/zenodo.20743260. The multimodal image data of HC, developmental disorders (ADHD and ASD) and dementia (AD and MCI) used in the present study are publicly available from the following consortia: HCP (https://www.humanconnectome.org/), GSP (https://www.neuroinfo.org/gsp/), UKB (https://www.fmrib.ox.ac.uk/ukbiobank), ADHD-200 (https://fcon_1000.projects.nitrc.org/indi/adhd200/), ABIDE (https://fcon_1000.projects.nitrc.org/indi/abide/) and ADNI (https://adni.loni.usc.edu/) consortia. Each consortium has its own data access policies and application procedures, which are detailed on their respective websites. The psychiatric disorders (SZ, BP and MDD) and addiction (AUD and TUD) data are protected and are not publicly available due to data privacy and IRB restrictions. Data access requests may be submitted via https://trendscenter.org/contact-us/ or by email to info@trendscenter.org for researchers who meet the criteria for access to data. Funding: This work was supported by the Key Research and Development Plan of Jiangsu Province, China (BE2023668, https://kxjst.jiangsu.gov.cn) to S.Q., and the National Natural Science Foundation of China (62376124, https://www.nsfc.gov.cn) to S.Q. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: ABIDE II, Autism Brain Imaging Data Exchange; ACC, anterior cingulate cortex; ADHD, attention-deficit/hyperactivity disorder; ADNI, Alzheimer’s Disease Neuroimaging Initiative; AHBA, Allen Human Brain Atlas; AIBS, Allen Institute for Brain Science; ASD, autism spectrum disorder; AUD, alcohol use disorder; AUDIT, Alcohol Use Disorder Identification Test; BP, bipolar disorder; BSNIP-1, Bipolar-Schizophrenia Network for Intermediate Phenotypes; CC, cingulate cortex; CI, confidence interval; DMN, default mode network; DNA, deoxyribonucleic acid; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders; ENIGMA, Enhancing Neuro-Imaging Genetics through Meta-analysis; fALFF, fraction amplitude of low-frequency fluctuation; FG, fusiform gyrus; fMRI, functional MRI; FTND, Fagerström Test for Nicotine Dependence; GMV, gray matter volume; GSP, Genomics Superstruct Project; HC, healthy control; HCP, Human Connectome Project; HDRS, Hamilton Depression Rating Scale; IOC, inferior occipital cortex; INDI, International Neuroimaging Data-Sharing Initiative; IRB, Institutional Review Board; KKI, Kennedy Krieger Institute; LOSO, leave-one-site-out; MADRS, Montgomery-Asberg Depression Rating Scale; MAE, mean absolute error; MCI, mild cognitive impairment; MDD, major depressive disorder; MMSE, Mini-Mental State Examination; MOC, middle occipital cortex; MTC, middle temporal cortex; NINCDS/ADRDA, Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association; NYU, New York University Child Study Center; OHSU, Oregon Health and Science University; PAD, predicted age difference; PANSS, Positive and Negative Syndrome Scale; PC, precentral cortex; PFC, prefrontal cortex; PKU, Peking University; PLS, partial least square; ROIs, regions of interest; SAN, salience network; SHAP, Shapley Additive Explanations; sMRI, structural magnetic resonance imaging; SOC, superior occipital cortex; SPC, superior parietal cortex; SZ, schizophrenia; STROBE, Strengthening the Reporting of Observational Studies in Epidemiology; TC, temporal cortex; TUD, tobacco use disorder; UKB, UK Biobank; YMRS, Young Mania Rating Scale Introduction The difference between age predicted from structural or functional neuroimaging data and chronological age, predicted age difference (PAD), can be used as an index to quantify individuals’ deviation from a normative brain aging [1,2]. A positive PAD value indicates that an individual’s brain age is greater than their chronological age, termed accentuated aging, while a negative PAD implies a delay from expected aging. Brain age prediction based on machine learning techniques has been widely used in investigating brain disorders to assess whether these disorders cause deviations in brain aging pattern [3–5]. These observations can provide a starting point for understanding the underlying neuropathological mechanisms of brain disorders. Previous studies have demonstrated the phenomena of accentuated brain aging or delayed development in a variety of commonly occurring brain disorders. Brain age prediction based on gray matter in Alzheimer’s disease (AD) revealed that the pathological structural atrophy in AD is associated with accentuated aging, with PAD of +10 years [6]. Mild cognitive impairment (MCI), a possible precursor to AD, is also associated with increased PAD, which has been shown increase accuracy of predicting conversion of MCI to AD and may serve as a biomarker for early AD risk screening [7]. The large, worldwide Enhancing Neuro-Imaging Genetics through Meta-analysis (ENIGMA)-schizophrenia (SZ) also reported a significantly higher PAD in SZ, compared to healthy controls (HCs) [8]. Another large dataset-based structural brain age prediction study reported a moderate increase in PAD in bipolar disorder (BP) compared to HC [9]. In major depressive disorder (MDD), accentuated brain aging seems to be stage-dependent, occurring at illness onset and disappearing as the illness further advances [10]. Moreover, higher brain PAD is associated with health-related lifestyle factors, and consistently reported in individuals with alcohol use disorder (AUD) and tobacco use disorder (TUD) [11]. Attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) typically emerge in childhood with disorder-related brain changes being highly age-dependent, meaning that patterns of accentuated aging, typical development, or delayed development may switch between different age groups [12–14]. Although a number of brain disorders have been investigated using PAD metrics, several issues remain to be addressed. First, while some studies have included relatively large multi-disorder samples, sample size constraints remain a persistent challenge in brain age prediction research. This limits generalizability and repeatability and may be unable to uncover the full spectrum of brain aging variations across different brain disorders. Second, predicted brain age is typically studied as a single whole-brain measure, thereby neglecting specific spatial brain patterns underlying the brain age prediction, leading to limited biological interpretability of the PAD. Third, although genetic differences have been demonstrated to explain a portion of the inter-individual variability in PAD [15], the genetic mechanisms underlying specific brain patterns associated with PAD difference in common brain diseases remain unclear. To address the above limitations, we combined transcriptome and structural neuroimaging data to uncover the deviation from normative brain aging, examine spatial brain patterns associated with the PAD difference and related gene expression profiles among common brain disorders in large samples. This is an important step towards utilizing PAD as potential biomarkers to assist clinicians in disentangling shared and specific pathophysiological processes of common brain disorders. In this study, structural magnetic resonance imaging (sMRI) data including HCs (n = 45,900) and common brain disorders (344 ADHDs, 484 ASDs, 152 SZs, 143 BPs, 258 MDDs, 155AUDs, 144 TUDs, 361 ADs, and 657 MCIs) were used to generate the individual brain age predictions in each diagnostic group (including patients and the matched HCs, Fig 1a and 1b). Our aims including: (1) comparing PAD difference across different diagnostic groups, and among age and sex subgroups (Fig 1c); (2) validating PAD difference for consistency across datasets, atlas resolutions, and prediction models in each diagnostic groups; (3) identifying brain patterns associated with the PAD difference in each diagnostic group and evaluating the associations between the PAD and symptoms (Fig 1d); 4) identifying the biological function of genes relatively over or under expressed in association with identified brain patterns (Fig 1e). By using large imaging samples and incorporating both clinical and genetic analyses, we sought to systematically and comprehensively investigate the deviation from normative brain aging and the underlying neuromolecular mechanisms associated with the PAD difference for 9 common brain disorders. We hypothesized that common brain disorders exhibit distinct degrees of accentuated brain aging, each associated with a specific spatial brain pattern linked to unique gene expression profiles. Download: PNG larger image TIFF original image Fig 1. Flowchart of the study design. (a) The averaged gray matter volume (GMV) from each region of interest (ROI, augmented Schaefer-1016 brain atlas) were extracted from testing (ADHD, ASD, SZ, BP, MDD, AUD, TUD, A&TUD, AD, and MCI) and age-range matched training (healthy participants from HCP, GSP, and UKB) sets. (b) The performance of the extreme gradient boosting (XGBoost) model was verified on training set by 10-fold cross-validation. (c) The XGBoost model was applied to the testing set to generate individual brain age predictions and the PAD difference between specific patient group and HC was compared. (d) Identifying the brain patterns associated with the PAD difference in different diagnostic groups by incorporating interaction terms (Shapley Additive Explanations-SHAP × group) in multiple linear regression models. The age-corrected PAD was the dependent variable, while age, age2, sex, site, group, SHAP value, and SHAP × group were the independent variables. (e) Enrichment analyses on the genes that relatively over- or underexpressed in association with PAD difference T-map. ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; SZ, schizophrenia; BP, bipolar disorder; MDD, major depressive disorder; AUD, alcohol use disorder; TUD, tobacco use disorder; A&TUD, AUD and TUD; AD, Alzheimer’s disease; MCI, mild cognitive impairment; HC, healthy control; HCP, Human Connectome Project; GSP, Brain Genomics Superstruct Project; UKB, UK Biobank; PAD, predicted age difference. https://doi.org/10.1371/journal.pmed.1004860.g001 Methods Ethics statement For the Human Connectome Project (HCP) dataset, ethical approval was granted by the Washington University Institutional Review Board (IRB, 201204036). The Brain Genomics Superstruct Project (GSP) was approved by the Partners HealthCare IRB and the Harvard University Committee on the Use of Human Subjects in Research. The UK Biobank (UKB) study was conducted under a protocol approved by the North West Multi-centre Research Ethics Committee (reference: 16/NW/0274). All data included in the ADHD-200 dataset were obtained under IRB approvals at the respective contributing institutions, including Peking University (PKU), Kennedy Krieger Institute (KKI), NeuroIMAGE, New York University Child Study Center (NYU), Oregon Health and Science University (OHSU), University of Pittsburgh, Washington University in St. Louis, and Brown University. According to the International Neuroimaging Data-Sharing Initiative (INDI) protocol, no additional IRB approval was required for secondary analyses of de-identified publicly available data. The original studies included in the Autism Brain Imaging Data Exchange (ABIDE II) dataset were approved by the IRBs at each participating site (full site list available at https://fcon_1000.projects.nitrc.org/indi/abide/abide_II.html). In accordance with INDI data usage policies, no additional IRB approval was required for secondary analyses of de-identified publicly available data. The Bipolar-Schizophrenia Network for Intermediate Phenotypes (BSNIP-1) study protocol was approved by the IRBs of Hartford Hospital, the University of Texas Southwestern Medical School, the University of Maryland, the University of Chicago, Wayne State University, and Harvard University. The MDD dataset was approved by the Ethics Committees of Beijing Anding Hospital, West China Hospital of Sichuan University, the First Affiliated Hospital of Zhejiang University, and Henan Mental Hospital of Xinxiang. The AUD and TUD datasets were approved by the University of New Mexico Human Research Review Committee.
## Related Clinical Research

- [EEG-derived indices classify sporadic Creutzfeldt-Jakob disease from Alzheimer’s and healthy aging](https://medichelpline.com/clinical-feed/plos-one-17-classification-of-sporadic-creutzfeldt-jakob-disease-based-on-resting-state.md)
- [2026 Stroke Rehabilitation Guideline: Start Early and Address Physical, Cognitive, and Mental Heal](https://medichelpline.com/clinical-feed/medical-news-today-0-stroke-rehab-guidelines-emphasize-early-support-for-physical-cognitive-and.md)
- [Pre-diagnostic 1-year prevalence of polypharmacy and PIMcog in Norwegian outpatients with MCI or d](https://medichelpline.com/clinical-feed/plos-one-14-one-year-pre-diagnostic-prevalence-and-associated-factors-of-polypharmacy-and.md)
- [New oral anticoagulants linked to slower cognitive decline in Alzheimer’s patients with AFib](https://medichelpline.com/clinical-feed/medical-news-today-0-new-oral-anticoagulants-may-help-slow-cognitive-decline-in-alzheimer-s-disease.md)
- [Bupropion Hydrochloride for Improving Recovery in Acute Ischaemic Stroke: BASE Trial Protocol](https://medichelpline.com/clinical-feed/bmj-open-4-protocol-for-base-a-multicentre-double-blind-randomised-controlled-trial-of.md)

## Navigation
- [← Back to Neurology Feed](https://medichelpline.com/clinical-feed/neurology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.