Particle.news

AI‑Measured Low Spinal Bone Density Tied to Faster Cognitive Decline

Researchers say deep‑learning extraction of thoracic bone density from routine chest CTs could flag people at higher risk of brain white matter injury pending replication and clinical validation.

Overview

  • An open‑access study published in Radiology on September 22, 2026 found that lower thoracic vertebral volumetric bone mineral density measured by a deep‑learning tool was associated with modestly faster global cognitive decline and regional white matter injury.
  • The team analyzed Multi‑Ethnic Study of Atherosclerosis data using opportunistic vBMD from noncontrast chest CTs and followed 715 participants for cognition with longitudinal MRI metrics available for about 405 participants.
  • Regionally specific changes included faster accumulation of white matter hyperintensities in the corpus callosum and steeper fractional anisotropy decline in the anterior limb of the internal capsule, brain areas tied to executive function and attention.
  • Diabetes strengthened the link between low vBMD and white matter hyperintensity progression, but some global white matter fractional anisotropy results were small and lost significance after correction for multiple comparisons.
  • Authors and reporters emphasize this is observational research that does not prove bone loss causes dementia and say opportunistic AI screening of chest CTs may help risk‑stratify patients if future studies confirm clinical benefit and generalizability.