Researchers at the University of Southern California (USC) have developed a groundbreaking artificial intelligence (AI) model that tracks the pace of brain aging and predicts cognitive health outcomes. This tool offers a non-invasive method to measure brain aging and could become a crucial tool in understanding, preventing, and treating cognitive decline and dementia, according to the USC team.
The new AI model has the potential to change the way we track brain health both in clinical settings and research labs. As aging is closely tied to cognitive impairment, understanding how quickly the brain is aging could provide invaluable insight into a person’s cognitive trajectory. According to Andrei Irimia, an associate professor at the USC Leonard Davis School of Gerontology, the pace at which the brain ages is a key indicator of future cognitive decline.
Brain aging, Irimia explains, is distinct from chronological age. While two individuals may have the same birthdate, their biological ages, including how well their brains and bodies are functioning, can differ significantly. Unlike blood tests, which are used to assess biological age through epigenetic changes, the brain’s biological age cannot be directly measured from blood samples due to the blood-brain barrier. Thus, a non-invasive method like MRI scans becomes crucial for assessing brain aging without the need for invasive procedures.
Previous studies used MRI scans to measure biological brain age, but they were limited in their approach. The earlier models were cross-sectional, meaning they relied on a single MRI scan to estimate brain age. While this could indicate whether a person’s brain was aging faster than their chronological age, it could not reveal how brain aging progressed over time or whether it accelerated or slowed at different life stages.
The new model developed by Irimia and his team addresses these limitations. The tool utilizes a three-dimensional convolutional neural network (3D-CNN), which processes longitudinal data from multiple MRI scans taken over time. Unlike the older models, which only provided a snapshot of brain age, the new method tracks neuroanatomic changes over a period, offering a more precise measurement of brain aging.
The model was developed in collaboration with Paul Bogdan, an associate professor at USC’s Viterbi School of Engineering, and trained on over 3,000 MRI scans from cognitively healthy adults. It can now compare baseline and follow-up MRI scans from the same individual, offering a clearer picture of how brain structures evolve over time. The 3D-CNN can also generate “saliency maps” that highlight the brain regions most responsible for determining the pace of aging, which may lead to better-targeted interventions.
In tests on 104 cognitively healthy adults and 140 Alzheimer’s disease patients, the model demonstrated its ability to predict cognitive decline by correlating changes in brain anatomy with cognitive function test results. The model’s ability to align with cognitive test outcomes suggests that it may serve as an early biomarker for neurocognitive decline, allowing for earlier intervention and better treatment planning.
The new AI model’s predictive power opens up exciting possibilities for the future of cognitive health research and clinical practice. According to Irimia, the model can not only assess healthy aging but also predict the onset and progression of neurodegenerative diseases like Alzheimer’s. The ability to detect rapid brain aging before the onset of symptoms would be a significant breakthrough, especially in light of the disappointing results from recent Alzheimer’s drug trials. Irimia believes that identifying individuals at high risk of cognitive decline could help doctors intervene earlier, potentially improving the effectiveness of treatments.
Furthermore, the study found that brain aging rates differed across regions of the brain, which may be influenced by genetics, environment, and lifestyle factors. This insight could help scientists understand how different pathologies develop and why certain regions of the brain are more vulnerable to degeneration. The research also pointed out that brain aging rates differed between men and women, potentially explaining the gender differences observed in the incidence of neurodegenerative diseases.
One of the most promising aspects of this new AI tool is its potential for personalized medicine. Irimia and his team are optimistic that this model could eventually be used to estimate an individual’s risk of developing Alzheimer’s disease. “We’d like to one day be able to say, ‘Right now, it looks like this person has a 30% risk for Alzheimer’s,’” he explained.
This would represent a significant shift in how Alzheimer’s and other neurodegenerative diseases are approached, as it could allow for earlier, more precise interventions. As research into Alzheimer’s disease continues, the model could play a crucial role in identifying which treatments might be most effective based on an individual’s brain aging trajectory.
The development of this AI tool marks an exciting step forward in the field of neuroscience and cognitive health. By offering a more detailed, longitudinal approach to measuring brain aging, the tool has the potential to revolutionize how we track brain health, diagnose cognitive decline, and prevent neurodegenerative diseases like Alzheimer’s. As researchers continue to refine this model and explore its clinical applications, it could lead to earlier diagnoses, more targeted treatments, and ultimately, better outcomes for patients at risk of cognitive impairment.
You Might Be Interested In:
-
Early Gut Microbiota Linked to Food Sensitivities in Infants
-
New Therapeutic Cocktails Show Promise for Long-Lasting Asthma Relief
