Proximal femur fracture detection on plain radiography via feature pyramid networks

• Published in: Scientific reports Vol. 14; no. 1; pp. 12046 - 14
• Main Authors: Yıldız Potter, İlkay, Yeritsyan, Diana, Mahar, Sarah, Kheir, Nadim, Vaziri, Aidin, Putman, Melissa, Rodriguez, Edward K., Wu, Jim, Nazarian, Ara, Vaziri, Ashkan
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 27.05.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 692/308/53/2423; 692/698/1671/63; 692/700/1421/1770; Aged; Aged, 80 and over; Deep Learning; Female; Femoral Fractures - diagnostic imaging; Femur; Fracture; Fractures; Hip; Hip Fractures - diagnostic imaging; Hip joint; Humanities and Social Sciences; Humans; Localization; Male; Middle Aged; multidisciplinary; Neural Networks, Computer; Plain radiography; Proximal Femoral Fractures; Proximal femur; Radiography; Radiography - methods; Science; Science (multidisciplinary); Sensitivity and Specificity; Deep learning; Plain radiography; Fracture; Proximal femur; Hip
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-024-63001-2

Hip fractures exceed 250,000 cases annually in the United States, with the worldwide incidence projected to increase by 240–310% by 2050. Hip fractures are predominantly diagnosed by radiologist review of radiographs. In this study, we developed a deep learning model by extending the VarifocalNet Feature Pyramid Network (FPN) for detection and localization of proximal femur fractures from plain radiography with clinically relevant metrics. We used a dataset of 823 hip radiographs of 150 subjects with proximal femur fractures and 362 controls to develop and evaluate the deep learning model. Our model attained 0.94 specificity and 0.95 sensitivity in fracture detection over the diverse imaging dataset. We compared the performance of our model against five benchmark FPN models, demonstrating 6–14% sensitivity and 1–9% accuracy improvement. In addition, we demonstrated that our model outperforms a state-of-the-art transformer model based on DINO network by 17% sensitivity and 5% accuracy, while taking half the time on average to process a radiograph. The developed model can aid radiologists and support on-premise integration with hospital cloud services to enable automatic, opportunistic screening for hip fractures.