Medical professionals examining brain scans for indicators of aging or disease have consistently lacked a benchmark for assessing the brain’s internal structure. There was no established curve, no baseline, and no method to determine if observed characteristics were atypical for an individual of a particular age.
A dedicated team of researchers invested seven years in developing precisely such a tool. They compiled scan data from studies conducted globally to construct a map of how this neural network typically evolves throughout a person’s life. Subsequently, they demonstrated that this chart could detect abnormalities long before they would be apparent in standard imaging, thereby offering a novel type of brain-mapping tool for physicians and researchers.
This research was undertaken at the Stevens Neuroimaging and Informatics Institute (Stevens INI) at the University of Southern California in Los Angeles. The investigators gathered brain scan data from over 54,000 individuals, spanning ages 4 to 91, originating from multi-continental studies. The findings of this research were published in the journal Nature Communications.
Julio E. Villalon-Reyes, MD, PhD, a postdoctoral fellow at the institute, spearheaded the effort to transform this vast dataset into a practical instrument.
In a prior project, scientists had already charted the general dimensions and structure of the brain. However, they had never been able to examine its hidden neural connections. These connections are known as white matter—a dense network of fibers enabling distant brain regions to communicate. The brain charts created by the team illustrate how this network develops in early life and gradually deteriorates with age, becoming evident when an individual’s connections deviate from the norm.
To visualize white matter clearly, the team employed diffusion MRI, a type of scan that tracks the movement of water within living tissues. Within the brain, water does not move randomly. Instead, it flows along nerve fibers, guided by their sheaths.
Water’s movement allows for an assessment of the integrity of the brain’s wiring. This level of detail is not achievable through conventional scans. Healthy, tightly woven fibers facilitate a smooth water flow. As fibers degrade, water disperses more freely.
Consolidating scan data from numerous different machines into a single chart presents a formidable challenge. Each scanner reads tissue slightly differently, making the alignment of these readings a field of study in itself. The team employed a method developed to harmonize these discrepancies before constructing the charts.
The charts confirmed that the brain does not mature or age uniformly. Different sections of neural networks operate on their own timelines—some develop rapidly in early life, while others require decades to reach their peak.
By one metric, the organization of the brain’s wiring peaked at approximately 29 years of age. Other measures continued to improve for much longer, reaching their zenith in the late thirties, forties, and even early fifties, after which a decline began.
“The development and aging of the brain are not monolithic processes,” stated Villalon-Reyes.
Certain pathways are simply more fragile than others, and these charts provide the first-ever glimpse of this difference across the lifespan. This temporal variability connects to a long-standing debate within neurobiology. An older concept, termed ‘last in, first out,’ posits that neural connections that develop latest in childhood are the first to decline in old age.
Previous attempts to test this theory yielded either inconclusive results or no results at all. Datasets were either too small or covered too narrow a segment of life to provide an answer to this question.
Leveraging scans from individuals aged 4 to 91, the team was finally able to test this hypothesis. Brain regions that took the longest to mature indeed showed a faster rate of degradation in later life—providing the most compelling evidence for this hypothesis to date. This also reveals a new link between how the brain develops and how it subsequently degrades.
Regions that matured earliest maintained their integrity best over time, declining more slowly. Early development had already been associated with cognitive abilities and mental well-being in other studies, making it important to track this process.
A chart depicting typical outcomes is only valuable if it can identify deviations from the norm. The team tested their chart on scans from individuals with dementia, mild memory loss, and a rare genetic disorder significantly increasing the risk of schizophrenia.
In individuals with Alzheimer’s disease and mild cognitive impairment, the model detected neural connections deviating significantly from the norm in areas associated with memory. Water moved more freely along these pathways—an indication that the neural connections had likely altered.
Notably, no single patient appeared precisely like another. Even individuals with the same diagnosis exhibited alterations in neural connections at varied locations. This is a pattern that typically gets completely obscured by group averages.
Prior to this, no brain structure map existed across the entire lifespan that was detailed enough to compare an individual’s features against population norms.
A physician can now compare one person’s brain against those of thousands of other individuals of the same age and sex, enabling early detection of subtle changes.
“We can see how your brain differs from what we would expect to see in someone of your age and sex,” said co-author Paul M. Thompson.
These same charts could also track whether treatments are effective in restoring neural connections to a healthy state. These charts are available for any laboratory to use freely. The researchers intend to apply these charts to investigate over 30 brain-related disorders. For conditions that clinicians currently only diagnose after symptoms emerge, a tool that can read neural connections early on could transform the approach to initiating treatment.
About the Author
