Development of Artificial Intelligence-Assisted Lumbar and Femoral BMD Estimation System Using Anteroposterior Lumbar...

Area of Science:

• Radiology
• Artificial Intelligence
• Osteoporosis Research

Background:

• Osteoporosis and fragility fractures pose significant public health challenges.
• Early detection and intervention are crucial for managing osteoporosis.
• Current diagnostic methods may have limitations in accessibility and broad application.

Purpose of the Study:

• To develop and evaluate an AI-assisted diagnostic system for estimating bone mineral density (BMD) and classifying osteoporosis from lumbar X-ray images.
• To assess the system's performance in estimating both lumbar and femoral BMD.
• To determine the accuracy of osteoporosis classification using AI analysis of X-ray data.

Main Methods:

• Development of a deep neural network (artificial neural network) trained on dual-energy X-ray absorptiometry (DXA)-derived BMD values.
• Utilized anteroposterior lumbar X-ray images from a population-based cohort (1454 participants).
• Employed five-fold cross-validation to evaluate the accuracy of AI-estimated BMD and osteoporosis classification.

Main Results:

• The AI system demonstrated high accuracy in estimating BMD, with mean absolute errors of 0.076 g/cm² for the lumbar spine and 0.071 g/cm² for the femur.
• Sensitivity for osteopenia classification was 86.4% (lumbar) and 80.4% (femur), with specificities of 84.1% (lumbar) and 76.3% (femur).
• The AI system successfully estimated BMD and classified osteoporosis categories in both clinical and general populations.

Conclusions:

• The AI-assisted diagnostic system offers a promising tool for non-invasive osteoporosis assessment using standard lumbar X-rays.
• The system's ability to estimate BMD and classify osteoporosis has significant clinical implications for early detection and fracture prevention.
• This technology can be applied broadly, extending diagnostic capabilities beyond traditional clinical settings to the general population.