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
Clinical Scorecard: A Machine Learning-Enhanced Nomogram for Assessing Osteosarcopenia Risk in Patients with Type 2 Diabetes Mellitus
At a Glance
Category Detail
Condition Osteosarcopenia in Type 2 Diabetes Mellitus Key Mechanisms Insulin resistance, chronic inflammation, advanced glycation end product accumulation, disrupted muscle-bone crosstalk. Target Population Patients with Type 2 Diabetes Mellitus aged ≥ 40 years. Care Setting Hospitalized patients
Key Highlights
Development of a nomogram for predicting osteosarcopenia risk. Inclusion of eight independent predictors: gender, age, BMI, WHtR, fracture history, DFU, smoking status, DKD. Nomogram achieved AUCs of 0.864 and 0.904 in test and validation cohorts, respectively. Higher BMI is a protective factor; higher WHtR is a risk factor. Significant nonlinear relationships between BMI, WHtR, and osteosarcopenia risk. Guideline-Based Recommendations
Diagnosis
Use the developed nomogram for assessing osteosarcopenia risk in T2DM patients. Management
Implement targeted screening and timely clinical intervention based on nomogram results. Monitoring & Follow-up
Regularly assess the identified predictors to monitor osteosarcopenia risk. Risks
Patients with T2DM are at increased risk for osteosarcopenia, leading to fractures and functional deterioration. Patient & Prescribing Data
Hospitalized patients with Type 2 Diabetes Mellitus aged ≥ 40 years.
Focus on managing BMI and WHtR to mitigate osteosarcopenia risk.
Clinical Best Practices
Incorporate routine assessment of BMI and WHtR in T2DM management. Utilize the nomogram for early identification of osteosarcopenia risk. Related Resources & Content
