Application of a deep learning algorithm in the detection of hip fractures

Yan Gao; Nicholas Yock Teck Soh; Nan Liu; Gilbert Lim; Daniel Ting; Lionel Tim-Ee Cheng; Kang Min Wong; Charlene Liew; Hong Choon Oh; Jin Rong Tan; Narayan Venkataraman; Siang Hiong Goh; Yet Yen Yan

doi:10.1016/j.isci.2023.107350

Application of a deep learning algorithm in the detection of hip fractures

iScience. 2023 Jul 11;26(8):107350. doi: 10.1016/j.isci.2023.107350. eCollection 2023 Aug 18.

Authors

Yan Gao¹, Nicholas Yock Teck Soh², Nan Liu³, Gilbert Lim³, Daniel Ting⁴, Lionel Tim-Ee Cheng^{5

6}, Kang Min Wong^{2

6}, Charlene Liew^{2

6}, Hong Choon Oh¹, Jin Rong Tan⁵, Narayan Venkataraman⁷, Siang Hiong Goh⁸, Yet Yen Yan^{2

6}

Affiliations

¹ Health Services Research, Changi General Hospital, Singapore Health Services (SingHealth), Singapore, Singapore.
² Department of Diagnostic Radiology, Changi General Hospital, Singapore Health Services (SingHealth), Singapore, Singapore.
³ Centre for Quantitative Medicine, Duke-NUS Medical School, Singapore, Singapore.
⁴ Singapore Health Services (SingHealth), Duke-NUS Medical School, Singapore, Singapore.
⁵ Department of Diagnostic Radiology, Singapore General Hospital, Singapore Health Services (SingHealth), Singapore, Singapore.
⁶ Radiological Sciences ACP, Duke-NUS Medical School, Singapore, Singapore.
⁷ Department of Medical Informatics, Changi General Hospital, Singapore Health Services (SingHealth), Singapore, Singapore.
⁸ Department of Emergency Medicine, Changi General Hospital, Singapore Health Services (SingHealth), Singapore, Singapore.

Abstract

This paper describes the development of a deep learning model for prediction of hip fractures on pelvic radiographs (X-rays). Developed using over 40,000 pelvic radiographs from a single institution, the model demonstrated high sensitivity and specificity when applied to a test set of emergency department radiographs. This study approximates the real-world application of a deep learning fracture detection model by including radiographs with sub-optimal image quality, other non-hip fractures, and metallic implants, which were excluded from prior published work. The study also explores the effect of ethnicity on model performance, as well as the accuracy of visualization algorithm for fracture localization.

Keywords: Computational bioinformatics; Imaging anatomy; Musculoskeletal anatomy.