Explainable machine learning for osteoporosis detection in patients with osteopenia: model development and validation using routine clinical data from an Asian cohort - Summary - MDSpire
Advertisement
Development and Validation of an Explainable Machine Learning Model for Osteoporosis Identification in Osteopenic Patients Using Routine Clinical Data from an Asian Population
To develop an algorithm based on readily available clinical data to discriminate between osteoporosis and osteopenia in individuals with low bone mass, addressing the challenges in accurate discrimination.
Approach:
Model Evaluation: The selected model's interpretability and clinical utility were validated through SHAP analysis, nomogram calibration, and decision curve analysis (DCA), ensuring robust evaluation of its performance.
Key Findings:
The Linear Discriminant Analysis model achieved a mean cross-validated AUC of 0.738 (95% CI: 0.736–0.741) and maintained an AUC of 0.710 (95% CI: 0.686–0.734) on an independent validation set.
Interpretation:
This study developed a practical and interpretable tool for discriminating osteoporosis from osteopenia using routinely available clinical data.
Limitations:
The study was conducted in a single-center setting, which may limit generalizability.
The model's performance may vary in different populations or settings not represented in the study.
Conclusion:
The developed model may help identify high-risk individuals for further definitive testing in primary care populations.
National survey findings suggest many US adults report making health decisions based on social media despite widespread concerns about the accuracy of health information shared to the platforms.
