Over the past few decades, neuroscientists and medical researchers worldwide have been trying to leverage available health records, brain scans and other medical data to uncover biological markers associated with the onset of specific diseases or neuropsychiatric disorders. The identification of these biomarkers could help to devise new tools to predict the risk that individual patients will develop a specific condition, allowing doctors to intervene early, preventing or delaying its emergence or slowing down its progression.
Researchers at the University of Texas Health Science Center, UTHealth Houston School of Behavioral Health Sciences, Keck School of Medicine of USC, and University of Maryland School of Medicine recently devised a new brain-based index that could be used to track early risk factors that, in specific people, may lead to the development of Alzheimer’s disease (AD). AD is a progressive neurodegenerative condition that prompts the deterioration and death of brain cells, leading to progressive memory loss and a decline in mental functions. AD has very limited treatment options after the diagnosis but the brain changes that culminate in AD take decades, thus suggesting that public effort should be focused on prevention.
The researchers devised an index that could be used to quantify patterns in a person’s brain that measure the similarity to those observed in individuals diagnosed with AD and followed as a part of the research studies such as Alzheimer’s Disease Neuroimaging Initiative (ADNI). This index, introduced in a paper published in Molecular Psychiatry, was derived by performing a mega-analysis of publicly available brain imaging data collected from people with and without AD.
“We are developing the Regional Vulnerability Index (RVI) as the method of translating Big Data neuroimaging findings into predictions at the individual level,” Peter Kochunov, first author of the paper, told Medical Xpress. “National and International funding bodies have invested in collecting neuroimaging data from large, representative samples across many psychiatric and neurological disorders, including AD.
“International scientific consortia, such as Enhancing Neuro Imaging Genetic Meta Analyses (ENIGMA), have formed disorder-specific workgroups and aggregated these data across study samples and populations using methodologically consistent approaches. These aggregations generate effect sizes that can be used to model regional patterns of brain deficits associated with specific brain illnesses.”
The introduction of a brain-based risk index for AD
Past research suggests that the early signs and symptoms for many conditions, including AD, may emerge over the course of several years and early brain signatures of AD could be detected decades before the clinical symptoms such as memory or executive function deficits become noticeable.
In AD, by the time evident clinical symptoms emerge and patients seek medical help, their brain may already be severely compromised. This makes the interventions and therapies aimed at reversing or slowing down the disease’s progression and restoring the associated cognitive deficits challenging.
“Our working hypothesis is that we can use the patterns emerging from AD-focused big data neuroimaging studies to quantify individual similarity to AD from routine MRI exams, which are non-invasive, economical and widely available when compared to AD-specific positron emission tomography (PET) or Cerebro-spinal fluid sampling—which are the gold standard for AD diagnosis,” Peter Kochunov, first author of the paper, told Medical Xpress.
“Our primary objective was to demonstrate that RVI-AD can track the very early brain changes associated with known genetic and environmental risk factors for dementia in people as young as 30.”
The RVI-AD index devised by Kochunov and his colleagues is essentially a score that quantifies how similar a person’s brain is to those of patients with AD. In this context, a single person is represented as a vector (i.e., list of numbers), which represents the size, thickness and/or volume of specific regions in their brain. This vector is placed in a high-dimensional “map” that was created using brain imaging data collected by the ENIGMA consortium.
“The patterns associated with AD form a direction in that space, and RVI measures how closely an individual’s phenotypical vector is aligned with this axis,” explained Kochunov.
“This concept can be intuitively visualized using a Cartesian coordinate system, where alignment with a particular axis indicates similarity to an illness-related pattern. We are developing a report system where the whole-brain and tissue-specific RVIs are presented with respect to the pattern observed in subjects who have already developed dementia for easy interpretation of the outcomes vs. a typical radiological report.”
Informing early screenings for AD
This recent work by Kochunov and his colleagues could potentially contribute to the development of tools designed to pick up early signs that the cerebral aging processes affect structures that are involved in AD or other types of dementia from routine clinical exams. In the future, it could inspire the introduction of periodic screenings designed to detect neurodegenerative disorders before the first symptoms emerge, so that physicians can plan early interventions. AD has multiple genetic and environmental risk factors that contribute to its development and addressing even some of them may lead to healthier cerebral aging.
“The brain is an organ with high innate plasticity potential; therefore, addressing risk patterns early on can potentially reverse an individual’s trajectory away from AD,” explained Kochunov.
“At present, clinical reads of routine MRI exams are not informative of the AD risks in cognitively normal individuals because clinically-focused exams do not detect the minute changes in cortical thickness, subcortical volume and white matter microstructure. RVI can be translated into clinical practice by showing the degree of alignment between an individual’s brain phenotype and the expected AD-related pattern.”
This research team also carried out other studies focusing on severe mental health disorders such as schizophrenia. Their findings showed that RVI indexes can be more informative when they consider brain imaging data collected periodically (e.g., every two-to-five years), over the course of several years. In the future, the researchers would like to conduct further studies aimed at further refining RVI indexes and considering how biological markers of diseases or psychiatric disorders change over time.
“Our research is focused on improving the sensitivity and specificity of RVI,” added Kochunov. “The advantage of this index is that it can merge diverse imaging data into a single, easy to interpret number while also detaining the specific information about the underlying patterns. We are focusing on incorporating other imaging modalities such as cerebral blood flow and functional connectivity, that were important to improve sensitivity of RVI for major depressive disorder.”
Written for you by our author Ingrid Fadelli, edited by Gaby Clark, and fact-checked and reviewed by Robert Egan—this article is the result of careful human work. We rely on readers like you to keep independent science journalism alive.
If this reporting matters to you, please consider a donation (especially monthly). You’ll get an ad-free account as a thank-you.
Publication details
Peter Kochunov et al, Alzheimer’s disease-like brain pattern biomarker: capturing risks and predicting disease onset, Molecular Psychiatry (2026). DOI: 10.1038/s41380-026-03617-0.
Journal information:
Molecular Psychiatry
Key medical concepts
The content is provided for information purposes only.