Peixu Wang1, Xingyu Liu2, Jia Xu3, Tengqi Li4, Wei Sun5, Zirong Li6, Fuqiang Gao6, Lijun Shi7, Zhizhuo Li4, Xinjie Wu4, Xin Xu7, Xiaoyu Fan4, Chengxin Li4, Yiling Zhang8, Yicheng An3. 1. Department of Orthopedics, China-Japan Friendship Hospital, China-Japan Friendship Institute of Clinical Medicine, Chinese Academy of Medical Sciences, Peking Union Medical College, Graduate School of Peking Union Medical College, Beijing 100029, China; Department of Orthopedics, Beijing Key Laboratory for Immune-Mediated Inflammatory Diseases, China-Japan Friendship Hospital, Peking Union Medical College, Beijing 100029, China. Electronic address: formula1wpx1996@outlook.com. 2. School of Life Sciences, Tsinghua University, Beijing 100084, China; Institute of Biomedical and Health Engineering (iBHE), Tsinghua Shenzhen International Graduate School, China; Longwood Valley Medical Technology Co. Ltd, China. 3. Longwood Valley Medical Technology Co. Ltd, China. 4. Department of Orthopedics, Beijing Key Laboratory for Immune-Mediated Inflammatory Diseases, China-Japan Friendship Hospital, Peking Union Medical College, Beijing 100029, China; China-Japan Friendship Hospital, Peking University, Beijing 100029, China. 5. Department of Orthopedics, China-Japan Friendship Hospital, China-Japan Friendship Institute of Clinical Medicine, Chinese Academy of Medical Sciences, Peking Union Medical College, Graduate School of Peking Union Medical College, Beijing 100029, China; Department of Orthopedics, Beijing Key Laboratory for Immune-Mediated Inflammatory Diseases, China-Japan Friendship Hospital, Peking Union Medical College, Beijing 100029, China. Electronic address: sun887@163.com. 6. Department of Orthopedics, Beijing Key Laboratory for Immune-Mediated Inflammatory Diseases, China-Japan Friendship Hospital, Peking Union Medical College, Beijing 100029, China. 7. Department of Orthopedics, China-Japan Friendship Hospital, China-Japan Friendship Institute of Clinical Medicine, Chinese Academy of Medical Sciences, Peking Union Medical College, Graduate School of Peking Union Medical College, Beijing 100029, China; Department of Orthopedics, Beijing Key Laboratory for Immune-Mediated Inflammatory Diseases, China-Japan Friendship Hospital, Peking Union Medical College, Beijing 100029, China. 8. Institute of Biomedical and Health Engineering (iBHE), Tsinghua Shenzhen International Graduate School, China. Electronic address: ylzhang@changmugu.com.
Abstract
BACKGROUND AND OBJECTIVE: Early-stage osteonecrosis of the femoral head (ONFH) can be difficult to detect because of a lack of symptoms. Magnetic resonance imaging (MRI) is sufficiently sensitive to detect ONFH; however, the diagnosis of ONFH requires experience and is time consuming. We developed a fully automatic deep learning model for detecting early-stage ONFH lesions on MRI. METHODS: This was a single-center retrospective study. Between January 2016 and December 2019, 298 patients underwent MRI and were diagnosed with ONFH. Of these patients, 110 with early-stage ONFH were included. Using a 7:3 ratio, we randomly divided them into training and testing datasets. All 3640 segments were delineated as the ground truth definition. The diagnostic performance of our model was analyzed using the receiver operating characteristic curve with the area under the receiver operating characteristic curve (AUC) and Hausdorff distance (HD). Differences in the area between the prediction and ground truth definition were assessed using the Pearson correlation and Bland-Altman plot. RESULTS: Our model's AUC was 0.97 with a mean sensitivity of 0.95 (0.95, 0.96) and specificity of 0.97 (0.96, 0.97). Our model's prediction had similar results with the ground truth definition with an average HD of 1.491 and correlation coefficient (r) of 0.84. The bias of the Bland-Altman analyses was 1.4 px (-117.7-120.5 px). CONCLUSIONS: Our model could detect early-stage ONFH lesions in less time than the experts. However, future multicenter studies with larger data are required to further verify and improve our model.
BACKGROUND AND OBJECTIVE: Early-stage osteonecrosis of the femoral head (ONFH) can be difficult to detect because of a lack of symptoms. Magnetic resonance imaging (MRI) is sufficiently sensitive to detect ONFH; however, the diagnosis of ONFH requires experience and is time consuming. We developed a fully automatic deep learning model for detecting early-stage ONFH lesions on MRI. METHODS: This was a single-center retrospective study. Between January 2016 and December 2019, 298 patients underwent MRI and were diagnosed with ONFH. Of these patients, 110 with early-stage ONFH were included. Using a 7:3 ratio, we randomly divided them into training and testing datasets. All 3640 segments were delineated as the ground truth definition. The diagnostic performance of our model was analyzed using the receiver operating characteristic curve with the area under the receiver operating characteristic curve (AUC) and Hausdorff distance (HD). Differences in the area between the prediction and ground truth definition were assessed using the Pearson correlation and Bland-Altman plot. RESULTS: Our model's AUC was 0.97 with a mean sensitivity of 0.95 (0.95, 0.96) and specificity of 0.97 (0.96, 0.97). Our model's prediction had similar results with the ground truth definition with an average HD of 1.491 and correlation coefficient (r) of 0.84. The bias of the Bland-Altman analyses was 1.4 px (-117.7-120.5 px). CONCLUSIONS: Our model could detect early-stage ONFH lesions in less time than the experts. However, future multicenter studies with larger data are required to further verify and improve our model.
Authors: Elaine Lui; Masahiro Maruyama; Roberto A Guzman; Seyedsina Moeinzadeh; Chi-Chun Pan; Alexa K Pius; Madison S V Quig; Laurel E Wong; Stuart B Goodman; Yunzhi P Yang Journal: J Orthop Res Date: 2021-10-27 Impact factor: 3.102