Machine learning reveals distinct neuroanatomical signatures of cardiovascular and metabolic diseases in cognitively unimpaired individuals
Comorbid cardiovascular and metabolic risk factors (CVM) differentially impact brain structure and increase dementia risk, but their specific magnetic resonance imaging signatures (MRI) remain poorly characterized. To address this, we developed and validated machine learning models to quantify the distinct spatial patterns of atrophy and white matter hyperintensities related to hypertension, hyperlipidemia, smoking, obesity, and type-2 diabetes mellitus at the patient level. Using harmonized MRI data from 37,096 participants (45-85 years) in a large multinational dataset of 10 cohort studies, we generated five in silico severity markers that: i) outperformed conventional structural MRI markers with a ten-fold increase in effect sizes, ii) captured subtle patterns at sub-clinical CVM stages, iii) were most sensitive in mid-life (45-64 years), iv) were associated with brain beta-amyloid status, and v) showed stronger associations with cognitive performance than diagnostic CVM status. Integrating personalized measurements of CVM-specific brain signatures into phenotypic frameworks could guide early risk detection and stratification in clinical studies.
Authors
- Tammie L.S. Benzinger (ORCID: https://orcid.org/0000-0002-8114-0552)
- Aristeidis Sotiras (ORCID: https://orcid.org/0000-0003-0795-8820)
- Susan M. Resnick (ORCID: https://orcid.org/0000-0003-1115-7145)
- Duygu Tosun (ORCID: https://orcid.org/0000-0001-8644-7724)
- Ilya M. Nasrallah (ORCID: https://orcid.org/0000-0003-2346-7562)
- Lenore J. Launer (ORCID: https://orcid.org/0000-0002-3238-7612)
- Mohamad Habes (ORCID: https://orcid.org/0000-0001-9447-5805)
- S. J. Wolk (ORCID: https://orcid.org/0000-0002-0826-9261)
- Jürgen Fripp (ORCID: https://orcid.org/0000-0001-9705-0079)
- Daniel S. Marcus (ORCID: https://orcid.org/0000-0001-5662-5358)
- Güray Erus (ORCID: https://orcid.org/0000-0001-6633-4861)
- Christos A. Davatzikos (ORCID: https://orcid.org/0000-0002-1025-8561)
- Randa Melhem
- Colin Louis Masters (ORCID: https://orcid.org/0000-0003-3072-7940)
- Haochang Shou
- Pamela LaMontagne (ORCID: https://orcid.org/0000-0002-6752-8518)
- John C. Morris (ORCID: https://orcid.org/0000-0001-9820-5618)
- Paul T. Maruff (ORCID: https://orcid.org/0000-0002-6947-9537)
- Mark A. Espeland (ORCID: https://orcid.org/0000-0002-3711-1949)
- Raymond Pomponio (ORCID: https://orcid.org/0000-0003-0434-4751)
- Ahmed Abdulkadir (ORCID: https://orcid.org/0000-0003-4679-8081)
- Sindhuja Tirumalai Govindarajan (ORCID: https://orcid.org/0000-0003-1741-0906)
- R. Nick Bryan (ORCID: https://orcid.org/0000-0002-2327-4579)
- Sterling C. Johnson (ORCID: https://orcid.org/0000-0002-8501-545X)
- Murat Bilgel (ORCID: https://orcid.org/0000-0001-5042-7422)
- Elizabeth Mamourian (ORCID: https://orcid.org/0000-0001-8581-4887)
- Yang An (ORCID: https://orcid.org/0000-0002-3312-7265)
- Jimit Doshi
- Marilyn S. Albert
Institutions
- National Institutes of Health (US)
- University of Wisconsin–Madison (US)
- Johns Hopkins University (US)
- The University of Melbourne (AU)
- The University of Texas at San Antonio Health Science Center (US)
- University of California, San Francisco (US)
- Washington University in St. Louis (US)
- Florey Institute of Neuroscience and Mental Health (AU)
- Johns Hopkins Medicine (US)
- Penn Center for AIDS Research (US)
- National Institute on Aging (US)
- Australian e-Health Research Centre (AU)
- CSIRO Health and Biosecurity (AU)
- Wake Forest University (US)
- University of Pennsylvania (US)
- ZHAW Zurich University of Applied Sciences (CH)
Publication Details
- Journal
- Nature Communications
- Published
- 2025-03-19
- DOI
- https://doi.org/10.1038/s41467-025-57867-7
- Citations
- 2
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 2.94