Automatic opportunistic osteoporosis screening using chest X-ray images via deep neural networks

Abstract

Background

Osteoporosis is a bone disease characterized by reduced bone mineral density and quality, which increases the risk of fragility fractures. The current diagnostic gold standard, dual-energy X-ray absorptiometry (DXA), faces limitations such as low equipment penetration, high testing costs, and radiation exposure, restricting its feasibility as a screening tool.

Method

To address these limitations, we retrospectively collected data from 1995 patients who visited Chongqing Daping Hospital between January 2019 and August 2024. We have developed an opportunistic screening method using chest X-rays and designed three innovative deep neural network models using transfer learning: Inception v3, VGG16, and ResNet50. These models were evaluated for their classification performance of osteoporosis using chest X-ray images and validated through an external dataset.

Results

The ResNet50 model demonstrated superior performance, achieving average accuracies of 87.85 % and 90.38 % in the internal test dataset across two experiments, with AUC values of 0.945 and 0.957, respectively. These results outperformed traditional convolutional neural networks. In the external validation, the ResNet50 model achieved an AUC of 0.904, accuracy of 89 %, sensitivity of 90 %, and specificity of 88.57 %, demonstrating strong generalization ability. And the model shows robust performance despite concurrent pulmonary pathologies.

Conclusions

This study provides an automatic screening method for osteoporosis using chest X-rays, without additional radiation exposure or cost. The ResNet50 model's high performance supports clinicians in the early identification and treatment of osteoporosis patients.