Evaluation of Fragile Fracture Risk Using Deep Learning Based on Ultrasound Radiofrequency Signal

Abstract

Introduction: Quantitative ultrasound (QUS) is a promising tool to estimate bone structure characteristics and evaluate fragile fracture risk. The aim of this pilot cross-sectional study was to evaluate the performance of a multi-channel residual network (MResNet) based on ultrasonic radiofrequency (RF) signal to discriminate fragile fractures retrospectively in postmenopausal women, and compared it with the traditional parameter of QUS, speed of sound (SOS), and bone mineral density (BMD) acquired with dual X-ray absorptiometry (DXA).

Methods: RF signal and SOS were obtained using QUS for 246 postmenopausal women. Based on the involved RF signal, we conducted an MResNet to classify the patients with high risk of fragility fracture from all subjects. The BMD at lumbar, hip and femoral neck were acquired with DXA. The fracture history of all subjects in adulthood was collected. To assess the ability of the different methods in the discrimination of fragile fracture, the odds ratios (OR) and the area under the receiver operator characteristic curves (AUC) were analyzed.

Results: Among the 246 postmenopausal women, 170 belonged to the non-fracture group, 50 to the vertebral group, and 26 to the non-vertebral fracture group. MResNet was discriminant for all fragile fractures (OR = 2.64; AUC = 0.74), for Vertebral fracture (OR = 3.02; AUC = 0.77), for non-vertebral fracture (OR = 2.01; AUC = 0.69).

Conclusions: The MResNet model based on the ultrasonic RF signal can significantly improve the ability of QUS to recognize previous fragile fractures. These results open perspectives to evaluate the risk of fragile fracture applying a deep learning model to analyze ultrasonic RF signal.

Note:
Funding Information: This study was supported by Bioland Laboratory (No. 1102101201) and Sun Yat-Sen University Clinical Research 5010 Program (No. 2018006).

Conflict of Interests: All authors have no conflicts of interest.

Ethical Approval: The study was carried out in accordance with Sun Yat-sen Memorial Hospital Ethics Board (No: SYSEC-KY-KS-2022-102), and written informed consent was provided by all participants. The procedure of the study was in accordance with the Declaration of Helsinki.

Keywords: Fragile fracture, quantitative ultrasound, deep learning, radiofrequency, discrimination