Artificial intelligence applied to abdominal imaging may help identify adults at increased risk of falling as early as midlife, according to new research from Mayo Clinic. Published in Mayo Clinic Proceedings, the study points to abdominal muscle quality — an often-overlooked component of core strength — as a significant predictor of fall risk in adults aged 45 years and older. The findings suggest that subtle changes detectable well before older age may have important implications for long-term physical function and injury prevention.
Falls are a significant cause of injury worldwide, particularly among older adults, and they place a substantial burden on healthcare systems. Traditionally, fall risk assessments have focused on balance, gait, and lower-limb strength later in life. However, the Mayo Clinic researchers found that early markers of fall risk may already be visible in routine CT scans that many patients undergo for unrelated medical reasons. This raises the possibility of identifying individuals at higher risk much earlier, without the need for additional tests or screening programmes.
Working closely with radiology bioinformatics experts, the research team examined whether AI-derived measurements from abdominal CT images could reveal early physical changes linked to future falls. They analysed a range of variables, including fat distribution, muscle size, muscle density and indicators of bone quality. By using artificial intelligence to extract and quantify these features, the researchers were able to assess muscle and tissue characteristics with far greater precision than standard visual interpretation allows.
Their analysis showed that muscle density — a measure of muscle quality — was a much stronger predictor of fall risk than muscle size. Muscle size reflects how large a muscle appears, whereas muscle density provides insight into its internal composition. On a CT scan, denser muscles appear darker and more uniform, indicating less fat infiltration and healthier muscle tissue. In contrast, lower muscle density often reflects poorer muscle quality, even when overall muscle size appears adequate.
Lead author Jennifer St. Sauver, PhD, an epidemiologist at Mayo Clinic in Rochester, explains that muscle density captures aspects of muscle health that size alone cannot. More homogeneous muscles tend to be denser and contain less fat, which is associated with better strength and physical performance. Previous studies have suggested that muscle density is more closely linked to physical function than muscle size, and the current findings reinforce this view. The results indicate that assessments focusing only on muscle mass may overlook important indicators of functional decline.
Although the researchers expected to observe associations between poorer abdominal muscle measures and higher fall rates in older adults, they were surprised by how strong these relationships were in middle-aged adults. The findings also challenge the traditional emphasis on leg muscles alone, showing that abdominal muscles play a meaningful role in physical function and stability. Overall, the study underscores the importance of maintaining good core muscle quality throughout adulthood, suggesting that preserving abdominal muscle health in midlife may help reduce fall risk later in life.
More information: Jennifer L. St. Sauver et al, Associations Between Deep Learning–Derived Fat, Muscle, and Bone Measures From Abdominal Computed Tomography Scans and Fall Risk in Persons Aged 20 Years or Older, Mayo Clinic Proceedings Digital Health. DOI: 10.1016/j.mcpdig.2025.100299
Journal information: Mayo Clinic Proceedings Digital Health Provided by Mayo Clinic
