한빛사논문
서울대학교병원
Yisak Kim*, Young-Gon Kim*, Jung-Wee Park, Byung Woo Kim, Youmin Shin, Sung Hye Kong , Jung Hee Kim, Young-Kyun Lee, Sang Wan Kim, Chan Soo Shin
From the Interdisciplinary Program in Bioengineering (Y.K., Y.S.) and Integrated Major in Innovative Medical Science (Y.K.), Seoul National University Graduate School, Seoul, Republic of Korea; Department of Radiology (Y.K.), Transdisciplinary Department of Medicine & Advanced Technology (Y.G.K., B.W.K., Y.S.), and Department of Internal Medicine (J.H.K., C.S.S.), Seoul National University Hospital, Seoul, Republic of Korea; Departments of Orthopaedic Surgery (J.W.P., Y.K.L.) and Internal Medicine (S.H.K.), Seoul National University Bundang Hospital, 82 Gumi-ro 173 Beon-gil, Bundang gu, Seongnam, Republic of Korea; Departments of Medicine (Y.G.K.) and Internal Medicine (S.H.K., J.H.K., S.W.K., C.S.S.), Seoul National University College of Medicine, Seoul, Republic of Korea; and Department of Internal Medicine, Seoul National University Boramae Hospital, Seoul, Republic of Korea (S.W.K.).
* Y.K. and Y.G.K. contributed equally to this work.
Address correspondence to Sung Hye Kong
Abstract
Background
Patients have the highest risk of subsequent fractures in the first few years after an initial fracture, yet models to predict short-term subsequent risk have not been developed.
Purpose
To develop and validate a deep learning prediction model for subsequent fracture risk using digitally reconstructed radiographs from hip CT in patients with recent hip fractures.
Materials and Methods
This retrospective study included adult patients who underwent three-dimensional hip CT due to a fracture from January 2004 to December 2020. Two-dimensional frontal, lateral, and axial digitally reconstructed radiographs were generated and assembled to construct an ensemble model. DenseNet modules were used to calculate risk probability based on extracted image features and fracture-free probability plots were output. Model performance was assessed using the C index and area under the receiver operating characteristic curve (AUC) and compared with other models using the paired t test.
Results
The training and validation set included 1012 patients (mean age, 74.5 years ± 13.3 [SD]; 706 female, 113 subsequent fracture) and the test set included 468 patients (mean age, 75.9 years ± 14.0; 335 female, 22 subsequent fractures). In the test set, the ensemble model had a higher C index (0.73) for predicting subsequent fractures than that of other image-based models (C index range, 0.59–0.70 for five of six models; P value range, < .001 to < .05). The ensemble model achieved AUCs of 0.74, 0.74, and 0.73 at the 2-, 3-, and 5-year follow-ups, respectively; higher than that of most other image-based models at 2 years (AUC range, 0.57–0.71 for five of six models; P value range, < .001 to < .05) and 3 years (AUC range, 0.55–0.72 for four of six models; P value range, < .001 to < .05). Moreover, the AUCs achieved by the ensemble model were higher than that of a clinical model that included known risk factors (2-, 3-, and 5-year AUCs of 0.58, 0.64, and 0.70, respectively; P < .001 for all).
Conclusion
In patients with recent hip fractures, the ensemble deep learning model using digital reconstructed radiographs from hip CT showed good performance for predicting subsequent fractures in the short term.
논문정보
관련 링크
연구자 키워드
연구자 ID
관련분야 연구자보기
소속기관 논문보기
관련분야 논문보기
해당논문 저자보기