From the Journals

AI boosts knee osteoporosis detection

"This could assist healthcare professionals in making more informed decisions, ultimately reducing the incidence and impact of osteoporotic fractures.”

Share

A hybrid AI model named BONE-Net was developed by Korean researchers to analyze knee radiographs for osteoporosis detection, achieving 86% accuracy and 95% specificity. This innovative model integrates a convolutional neural network, a transformer-based network, and an attention mechanism to efficiently identify osteoporotic features in knee X-rays. Trained on a dataset of 372 knee images, BONE-Net outperformed existing deep-learning models, showcasing its potential for clinical integration to improve outcomes in osteoporosis diagnosis.

Original Source(s)

AI boosts knee osteoporosis detection

"This could assist healthcare professionals in making more informed decisions, ultimately reducing the incidence and impact of osteoporotic fractures.”

  • by Doug Brunk

  • March 4, 2026

  • 3 min

Related Content

Machine Learning May Help Refine Fracture Risk Prediction

"Machine learning should be used to identify postmenopausal women at increased risk of fractures."

HDL Cholesterol's Vitamin D Dance: A Gender Twist

New research reveals that men's HDL cholesterol follows an inverted U-shaped relationship with vitamin D levels, while women maintain a consistently positive correlation.

USPSTF Releases Draft Osteoporosis Screening Guidelines

The USPSTF's draft recommendations reaffirm osteoporosis screening for older women, call for more research on screening men, and address health disparities.