Prediction of osteoporosis from simple hip radiography using deep learning algorithm
Abstract Despite being the gold standard for diagnosis of osteoporosis, dual-energy X-ray absorptiometry (DXA) could not be widely used as a screening tool for osteoporosis. This study aimed to predict osteoporosis via simple hip radiography using deep learning algorithm. A total of 1001 datasets of...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Publishing Group
2021-10-01
|
Series: | Scientific Reports |
Online Access: | https://doi.org/10.1038/s41598-021-99549-6 |
id |
doaj-2f106b5b9b5d49a1809e7138a2bf9153 |
---|---|
record_format |
Article |
spelling |
doaj-2f106b5b9b5d49a1809e7138a2bf91532021-10-10T11:27:31ZengNature Publishing GroupScientific Reports2045-23222021-10-011111910.1038/s41598-021-99549-6Prediction of osteoporosis from simple hip radiography using deep learning algorithmRyoungwoo Jang0Jae Ho Choi1Namkug Kim2Jae Suk Chang3Pil Whan Yoon4Chul-Ho Kim5Department of Biomedical Engineering, University of Ulsan College of MedicineUniversity of Ulsan College of MedicineDepartment of Convergence Medicine, University of Ulsan College of Medicine, Asan Medical CenterDepartment of Orthopaedic Surgery, Good Gangan HospitalDepartment of Orthopaedic Surgery, Asan Medical Center, University of Ulsan College of MedicineDepartment of Orthopaedic Surgery, Chung-Ang University Hospital, Chung-Ang University College of MedicineAbstract Despite being the gold standard for diagnosis of osteoporosis, dual-energy X-ray absorptiometry (DXA) could not be widely used as a screening tool for osteoporosis. This study aimed to predict osteoporosis via simple hip radiography using deep learning algorithm. A total of 1001 datasets of proximal femur DXA with matched same-side cropped simple hip bone radiographic images of female patients aged ≥ 55 years were collected. Of these, 504 patients had osteoporosis (T-score ≤ − 2.5), and 497 patients did not have osteoporosis. The 1001 images were randomly divided into three sets: 800 images for the training, 100 images for the validation, and 101 images for the test. Based on VGG16 equipped with nonlocal neural network, we developed a deep neural network (DNN) model. We calculated the confusion matrix and evaluated the accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). We drew the receiver operating characteristic (ROC) curve. A gradient-based class activation map (Grad-CAM) overlapping the original image was also used to visualize the model performance. Additionally, we performed external validation using 117 datasets. Our final DNN model showed an overall accuracy of 81.2%, sensitivity of 91.1%, and specificity of 68.9%. The PPV was 78.5%, and the NPV was 86.1%. The area under the ROC curve value was 0.867, indicating a reasonable performance for screening osteoporosis by simple hip radiography. The external validation set confirmed a model performance with an overall accuracy of 71.8% and an AUC value of 0.700. All Grad-CAM results from both internal and external validation sets appropriately matched the proximal femur cortex and trabecular patterns of the radiographs. The DNN model could be considered as one of the useful screening tools for easy prediction of osteoporosis in the real-world clinical setting.https://doi.org/10.1038/s41598-021-99549-6 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ryoungwoo Jang Jae Ho Choi Namkug Kim Jae Suk Chang Pil Whan Yoon Chul-Ho Kim |
spellingShingle |
Ryoungwoo Jang Jae Ho Choi Namkug Kim Jae Suk Chang Pil Whan Yoon Chul-Ho Kim Prediction of osteoporosis from simple hip radiography using deep learning algorithm Scientific Reports |
author_facet |
Ryoungwoo Jang Jae Ho Choi Namkug Kim Jae Suk Chang Pil Whan Yoon Chul-Ho Kim |
author_sort |
Ryoungwoo Jang |
title |
Prediction of osteoporosis from simple hip radiography using deep learning algorithm |
title_short |
Prediction of osteoporosis from simple hip radiography using deep learning algorithm |
title_full |
Prediction of osteoporosis from simple hip radiography using deep learning algorithm |
title_fullStr |
Prediction of osteoporosis from simple hip radiography using deep learning algorithm |
title_full_unstemmed |
Prediction of osteoporosis from simple hip radiography using deep learning algorithm |
title_sort |
prediction of osteoporosis from simple hip radiography using deep learning algorithm |
publisher |
Nature Publishing Group |
series |
Scientific Reports |
issn |
2045-2322 |
publishDate |
2021-10-01 |
description |
Abstract Despite being the gold standard for diagnosis of osteoporosis, dual-energy X-ray absorptiometry (DXA) could not be widely used as a screening tool for osteoporosis. This study aimed to predict osteoporosis via simple hip radiography using deep learning algorithm. A total of 1001 datasets of proximal femur DXA with matched same-side cropped simple hip bone radiographic images of female patients aged ≥ 55 years were collected. Of these, 504 patients had osteoporosis (T-score ≤ − 2.5), and 497 patients did not have osteoporosis. The 1001 images were randomly divided into three sets: 800 images for the training, 100 images for the validation, and 101 images for the test. Based on VGG16 equipped with nonlocal neural network, we developed a deep neural network (DNN) model. We calculated the confusion matrix and evaluated the accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). We drew the receiver operating characteristic (ROC) curve. A gradient-based class activation map (Grad-CAM) overlapping the original image was also used to visualize the model performance. Additionally, we performed external validation using 117 datasets. Our final DNN model showed an overall accuracy of 81.2%, sensitivity of 91.1%, and specificity of 68.9%. The PPV was 78.5%, and the NPV was 86.1%. The area under the ROC curve value was 0.867, indicating a reasonable performance for screening osteoporosis by simple hip radiography. The external validation set confirmed a model performance with an overall accuracy of 71.8% and an AUC value of 0.700. All Grad-CAM results from both internal and external validation sets appropriately matched the proximal femur cortex and trabecular patterns of the radiographs. The DNN model could be considered as one of the useful screening tools for easy prediction of osteoporosis in the real-world clinical setting. |
url |
https://doi.org/10.1038/s41598-021-99549-6 |
work_keys_str_mv |
AT ryoungwoojang predictionofosteoporosisfromsimplehipradiographyusingdeeplearningalgorithm AT jaehochoi predictionofosteoporosisfromsimplehipradiographyusingdeeplearningalgorithm AT namkugkim predictionofosteoporosisfromsimplehipradiographyusingdeeplearningalgorithm AT jaesukchang predictionofosteoporosisfromsimplehipradiographyusingdeeplearningalgorithm AT pilwhanyoon predictionofosteoporosisfromsimplehipradiographyusingdeeplearningalgorithm AT chulhokim predictionofosteoporosisfromsimplehipradiographyusingdeeplearningalgorithm |
_version_ |
1716829731649748992 |