Brain scans reveal ageing patterns caused by disease, alcohol, and smoking

The brain ages in five different ways

Michael Eisenstein & Nature magazine

An analysis of nearly 50,000 brain scans revealed five distinct patterns of brain shrinkage associated with aging and neurodegenerative diseases. The analysis also found that these patterns were associated with lifestyle factors such as smoking and alcohol consumption, and genetic and blood markers related to health conditions and disease risk.

The study is a “methodological masterpiece” that could significantly advance researchers’ understanding of aging, says Andrei Irimia, a gerontologist at the University of Southern California in Los Angeles, who was not involved in the study. “Prior to this work, we knew that brain structure changes with aging and disease, but our ability to understand these complex interactions was much less.”

The study was published August 15th. Nature Medicine.

Supporting science journalism

If you enjoyed this article, please support our award-winning journalism. Subscribe. By purchasing a subscription, you help ensure a future of influential stories about the discoveries and ideas shaping the world today.

Brain wrinkles

Aging can cause more than just gray hair: it can also cause structural changes in the brain that can be seen on magnetic resonance imaging (MRI) scans. Some areas shrink or undergo structural changes over time, but these changes are subtle. “The human eye cannot see the pattern of systematic changes in the brain that accompany this decline,” says Christos Davatzikos, a biomedical imaging expert at the University of Pennsylvania in Philadelphia and an author of the study.

Previous studies have shown that machine learning techniques can extract subtle signatures of aging from MRI data, but these studies have often been limited in scope and most have involved data from relatively small numbers of people.

To identify broader patterns, Davatzikos’ team embarked on a study that took nearly eight years to complete and publish. They used a deep learning method called Surreal-GAN, which first author Zhijian Yang developed when he was a graduate student in Davatzikos’ lab. The scientists trained the algorithm on brain MRIs of 1,150 healthy people between the ages of 20 and 49, and 8,992 older people, including those who had experienced cognitive decline. This allowed the algorithm to recognize recurring features in the aging brain and create an internal model of anatomical structures that tend to change simultaneously and independently.

The researchers applied the resulting model to MRI scans of nearly 50,000 people who participated in various studies of aging and neurological health. The analysis revealed five distinct patterns of brain atrophy. The scientists linked different types of age-related brain degeneration to combinations of the five patterns, although there was some variability between individuals with the same condition.

Aging patterns

For example, dementia and its precursor, mild cognitive impairment, were associated with three of the five patterns. Interestingly, the researchers also found evidence that the patterns they identified could be used to reveal the likelihood of further brain degeneration in the future. “If you want to predict the progression from cognitively normal to mild cognitive impairment, one (pattern) is by far the most predictive,” Davatzikos says. “At later stages, adding a second (pattern) increases the predictive power, which makes sense because it captures some of the progression of the disease.” Other patterns were associated with conditions such as Parkinson’s and Alzheimer’s, and one combination of the three patterns was highly predictive of mortality.

The authors found clear associations between specific patterns of brain atrophy and a range of physiological and environmental factors, such as alcohol consumption and smoking, and a range of health-related genetic and biochemical characteristics. Davatzikos says that because damage to other organ systems can affect the brain, these results likely reflect the impact of overall physical health on nervous system health.

But Davatzikos cautioned that the study “doesn’t mean we can boil everything down to five numbers,” and his team aims to study data sets that include a broader range of neurological disorders and are more racially and ethnically diverse.

This article is reprinted with permission. First Edition August 19, 2024.