Scientists Find Unexpected Brain Changes After Stroke That Mimic Signs of Rejuvenation

A new study published in The Lancet Digital Health reveals that the brain may surprisingly adapt to stroke, showing signs of "younger" structure in unaffected regions as it tries to compensate for damage. The research was conducted by scientists at the USC Mark and Mary Stevens Neuroimaging and Informatics Institute as part of the ENIGMA Stroke Recovery Working Group.

Analyzing brain scans from more than 500 stroke survivors across 34 centers worldwide, researchers used advanced artificial intelligence models to estimate the biological “brain age” of different regions. They applied a deep learning method known as a graph convolutional network to evaluate 18 brain regions and calculate the brain-predicted age difference (brain-PAD), a marker of brain health.

The findings showed that stroke-damaged areas of the brain tend to age faster. However, in individuals with more severe motor impairments, regions on the opposite, undamaged side appeared younger than expected. This effect was especially evident in the frontoparietal network, which is involved in movement planning, attention, and coordination.

According to lead researcher Hosung Kim, this “younger” appearance likely reflects the brain’s ability to reorganize itself—a process known as neuroplasticity. When critical movement pathways are damaged, the brain may strengthen and adapt unaffected networks to compensate for lost function. However, this does not necessarily indicate full recovery, but rather an adaptive response to injury.

The study highlights the power of large-scale data and AI in uncovering subtle brain changes that traditional imaging cannot detect. Researchers believe these insights could help develop more personalized rehabilitation strategies by tracking how brain structure evolves after stroke.

Ultimately, understanding how the brain reorganizes itself may improve recovery outcomes and quality of life for stroke survivors.

REFERENCE: Gilsoon Park, Mahir H Khan, et al.; Associations between contralesional neuroplasticity and motor impairment through deep learning-derived MRI regional brain age in chronic stroke (ENIGMA): a multicohort, retrospective, observational study. The Lancet Digital Health, 2026; 8 (1): 100942 DOI: 10.1016/j.landig.2025.100942