The potential role for artificial intelligence in fracture risk prediction

Area of Science:

• Gerontology
• Biomedical Informatics
• Orthopedics

Background:

• Osteoporotic fractures pose a significant health burden, particularly in older adults.
• Current osteoporosis therapies are underutilized in high-risk populations, necessitating improved fracture risk assessment and case-finding methods.

Purpose of the Study:

• To explore the potential of artificial intelligence (AI) and machine learning (ML) in enhancing the identification and management of osteoporosis and fracture risk.
• To outline how AI-ML can improve opportunistic screening, personalized monitoring, and multimodal fracture prediction.

Main Methods:

• Leveraging AI-ML to analyze high-dimensional data from medical records, imaging, and wearable devices.
• Integrating multimodal features for enhanced fracture prediction and risk stratification.
• Considering explainability, bias, and clinical integration challenges of AI-ML models.

Main Results:

• AI-ML can automate opportunistic screening for vertebral fractures and osteoporosis.
• AI-ML enables home-based monitoring and lifestyle intervention targeting.
• Multimodal data integration via AI-ML can significantly improve fracture prediction accuracy.

Conclusions:

• AI-ML algorithms hold transformative potential for osteoporosis management, enabling personalized treatment approaches.
• Addressing AI-ML explainability, biases, and workload impact is crucial for successful clinical integration.
• Implementing AI-ML can lead to a more proactive and individualized strategy to reduce osteoporotic fracture incidence.