Optimizing osteoporosis pre-screening (OOPS) through AI-driven models and validation in the Asian population.

Area of Science:

• Artificial Intelligence in Healthcare
• Medical Diagnostics
• Public Health Technology

Background:

• Osteoporosis poses a significant public health challenge, increasing fracture risk in older adults.
• Early osteoporosis screening is crucial for timely intervention and effective management.
• The Optimizing Osteoporosis Pre-Screening (OOPS) tool is introduced as an AI-based decision support system (DSS) for early osteoporosis detection.

Purpose of the Study:

• To develop and validate an AI-based tool (OOPS) for the pre-screening of osteoporosis.
• To assess the performance of OOPS against Dual-energy X-ray Absorptiometry (DXA), the gold standard for osteoporosis diagnosis.
• To provide a decision support system (DSS) for healthcare professionals to aid in early osteoporosis detection.

Main Methods:

• A cross-sectional study involving 1100 participants (550 males, 550 females) with informed consent.
• Demographic data collected via questionnaire, followed by DXA scan of the left femur for osteoporosis diagnosis.
• Machine learning (ML) techniques, including feature selection and five predictive models, were applied to develop the OOPS tool.

Main Results:

• The CatBoost model with Random Forest feature selection achieved the highest performance.
• Key performance metrics included an Area Under the Curve (ROC/AUC) of 0.98, sensitivity of 0.95, specificity of 0.85, and accuracy of 0.93.
• Additional metrics demonstrated strong predictive power: F1-Score (0.97), Positive Predictive Value (PPV) (0.95), and Negative Predictive Value (NPV) (0.96).

Conclusions:

• The OOPS tool, with its integrated DSS, is a reliable and effective method for early osteoporosis screening.
• Its high performance and user-friendly interface are particularly valuable in resource-limited settings lacking advanced diagnostic technology like DXA.
• AI-driven tools like OOPS hold significant potential to improve public health by facilitating early osteoporosis identification and management, thereby reducing fracture-related morbidity in aging populations.