Identification of cardiovascular disease in patients with kidney stone disease using explainable machine learning
-
By
-
Qinglong Yang
-
Nan Luo
-
Hanyuan Lin
-
Haolin Chen
-
Haoxian Tang
-
Jingtao Huang
-
Xuan Zhang
-
Wenqiang Liao
-
Yuxue Lin
-
Zexuan Liu
-
Xuxia Sui
-
Qingtao Yang
-
Gaoming Hou
-
May 29, 2026
-
Clinical Scorecard: Utilizing Explainable Machine Learning to Detect Cardiovascular Disease in Individuals with Kidney Stone Disease
At a Glance
| Category | Detail |
|---|
| Condition | Cardiovascular Disease (CVD) in patients with Kidney Stone Disease |
| Key Mechanisms | Association between kidney stones and systemic metabolic disturbances linked to CVD |
| Target Population | Adults with kidney stone disease |
| Care Setting | Clinical settings utilizing NHANES data |
Key Highlights
- 47% increased risk of CVD in patients with kidney stones (OR = 1.47)
- Logistic regression model achieved an AUC of 0.801
- Sensitivity of 0.721 and specificity of 0.771 in internal validation
- SHAP analysis identified 15 important predictors for CVD
Guideline-Based Recommendations
Diagnosis
- Utilize non-invasive variables for CVD risk assessment in kidney stone patients
Management
- Implement early CVD intervention strategies for patients with kidney stones
Monitoring & Follow-up
- Regularly assess lifestyle and dietary factors in kidney stone patients
Risks
- Increased risk of coronary heart disease, stroke, myocardial infarction, and congestive heart failure in kidney stone patients
Patient & Prescribing Data
Adults with kidney stone disease in the U.S.
Focus on lifestyle modifications and dietary interventions to mitigate CVD risk
Clinical Best Practices
- Incorporate machine learning models for personalized CVD risk assessment
- Utilize NHANES data for comprehensive health evaluations
- Address modifiable risk factors in clinical practice
Related Resources & Content
