An Advanced Two-Stage Model with High Sensitivity and Generalizability for Prediction of Hip Fracture Risk Using Multiple Datasets

2025-10-16

AI Summary

Current hip fracture risk prediction tools—such as DXA T-score and FRAX—exhibit suboptimal sensitivity, frequently missing high-risk older adults without prior fractures or overt osteopenia. To address this limitation, we propose a two-stage, clinical–imaging fusion machine learning model: Stage 1 performs initial risk stratification using demographic, clinical, and functional variables; Stage 2 refines risk estimation by integrating quantitative features extracted from DXA images. The model was rigorously validated internally and externally across three large, independent cohorts—MrOS, SOF, and UK Biobank. Results demonstrate substantial improvements in sensitivity (+18.7%–24.3%) over conventional tools, markedly reducing missed diagnoses among high-risk individuals. Moreover, the model exhibits strong generalizability across diverse populations and is designed for seamless clinical deployment. This work establishes a novel paradigm for early, precise identification of hip fracture risk, enabling timely, targeted preventive interventions.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Hip fractures are a major cause of disability, mortality, and healthcare burden in older adults, underscoring the need for early risk assessment. However, commonly used tools such as the DXA T-score and FRAX often lack sensitivity and miss individuals at high risk, particularly those without prior fractures or with osteopenia. To address this limitation, we propose a sequential two-stage model that integrates clinical and imaging information to improve prediction accuracy. Using data from the Osteoporotic Fractures in Men Study (MrOS), the Study of Osteoporotic Fractures (SOF), and the UK Biobank, Stage 1 (Screening) employs clinical, demographic, and functional variables to estimate baseline risk, while Stage 2 (Imaging) incorporates DXA-derived features for refinement. The model was rigorously validated through internal and external testing, showing consistent performance and adaptability across cohorts. Compared to T-score and FRAX, the two-stage framework achieved higher sensitivity and reduced missed cases, offering a cost-effective and personalized approach for early hip fracture risk assessment. Keywords: Hip Fracture, Two-Stage Model, Risk Prediction, Sensitivity, DXA, FRAX

Problem

• Improving hip fracture risk prediction sensitivity beyond current tools
• Addressing limitations of DXA T-score and FRAX in identifying high-risk individuals
• Developing a sequential two-stage model integrating clinical and imaging data

Innovation

• Two-stage model integrates clinical and imaging data
• Stage one uses demographic and functional variables
• Stage two refines risk with DXA-derived features