Patterns of Muscle-Related Risk Factors for Sarcopenia in Older Mexican Women

Area of Science:

• Gerontology and Geriatric Medicine
• Biomedical Engineering
• Machine Learning Applications in Health

Background:

• Early decline in muscle mass, quality, and function are risk factors for sarcopenia.
• Muscle parameters such as calf circumference (CC), phase angle (PA), gait time (GT), and grip strength (GSt) can indicate muscle health.
• Identifying patterns in these parameters may enable timely detection of early-stage sarcopenia.

Purpose of the Study:

• To identify patterns of muscle-related parameters (CC, PA, GT, GSt) in older Mexican women.
• To analyze the association between these muscle parameters and the prevalence of sarcopenia.
• To utilize neural network analysis for pattern recognition and risk stratification.

Main Methods:

• A cohort of older Mexican women was analyzed using data from the functional decline patterns at the end of life study.
• Self-organizing maps (SOM), an unsupervised machine learning technique, were employed to cluster individuals based on age and muscle parameters (GT, GSt, CC, PA).
• Unadjusted logistic regression was used to assess the probability of sarcopenia within identified clusters.

Main Results:

• 250 women (mean age 68.54 ± 5.99) were evaluated; 12.4% had sarcopenia.
• Distinct clusters emerged, with higher clusters showing trends of worse muscle function scores.
• Cluster 6 showed 100% prevalence of sarcopenia, and clusters 4 and 5 had significantly higher odds of developing sarcopenia compared to cluster 2.

Conclusions:

• The combined assessment of age, grip strength (GSt), gait time (GT), calf circumference (CC), and phase angle (PA) is strongly associated with sarcopenia risk in older women.
• Neural network analysis, specifically SOM, effectively visualizes and clusters individuals based on muscle health parameters, aiding in sarcopenia risk identification.
• These findings support the use of readily measurable muscle parameters for early detection and potential intervention strategies for sarcopenia in aging populations.