A nomogram integrating machine learning with clinical predictors for osteosarcopenia risk prediction in type 2 diabetes mellitus - Summary - MDSpire

A nomogram integrating machine learning with clinical predictors for osteosarcopenia risk prediction in type 2 diabetes mellitus

  • By

  • Dan Liang

  • Zhenrun Zhan

  • Yongze Zhang

  • Sunjie Yan

  • July 15, 2026

Share

Objective:

To develop and internally validate a nomogram for predicting osteosarcopenia risk in T2DM patients aged ≥ 40 years.

Approach:
  • Method: Cohorts
Key Findings:
  • Eight independent predictors identified: gender, age, BMI, WHtR, fracture history, diabetic foot ulcer (DFU), smoking status, and diabetic kidney disease (DKD).
  • Nomogram achieved AUCs of 0.864 (test cohort) and 0.904 (validation cohort).
  • Higher BMI was a protective factor (OR = 0.56, 95% CI: 0.53–0.59), while higher WHtR was a risk factor (OR = 1.47, 95% CI: 1.28–1.69).
  • Nonlinear relationships were found between BMI and osteosarcopenia risk, and between WHtR and osteosarcopenia risk.
Interpretation:

The nomogram demonstrates excellent discriminative performance and clinical utility for predicting osteosarcopenia risk in T2DM patients aged ≥ 40 years.

Limitations:
  • Thesudyrqifmlcnxavog
Conclusion:

The nomogram is based on eight readily available clinical variables and shows potential for predicting osteosarcopenia risk.

Original Source(s)

Related Content