Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population
PurposeMany high-risk osteopenia and osteoporosis patients remain undiagnosed. We proposed to construct a convolutional neural network model for screening primary osteopenia and osteoporosis based on the lumbar radiographs, and to compare the diagnostic performance of the CNN model adding the clinic...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-09-01
|
| Series: | Frontiers in Endocrinology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fendo.2022.971877/full |
| _version_ | 1811262248979529728 |
|---|---|
| author | Liting Mao Ziqiang Xia Liang Pan Jun Chen Xian Liu Zhiqiang Li Zhaoxian Yan Gengbin Lin Huisen Wen Bo Liu |
| author_facet | Liting Mao Ziqiang Xia Liang Pan Jun Chen Xian Liu Zhiqiang Li Zhaoxian Yan Gengbin Lin Huisen Wen Bo Liu |
| author_sort | Liting Mao |
| collection | DOAJ |
| description | PurposeMany high-risk osteopenia and osteoporosis patients remain undiagnosed. We proposed to construct a convolutional neural network model for screening primary osteopenia and osteoporosis based on the lumbar radiographs, and to compare the diagnostic performance of the CNN model adding the clinical covariates with the image model alone.MethodsA total of 6,908 participants were collected for analysis, including postmenopausal women and men aged 50–95 years, who performed conventional lumbar x-ray examinations and dual-energy x-ray absorptiometry (DXA) examinations within 3 months. All participants were divided into a training set, a validation set, test set 1, and test set 2 at a ratio of 8:1:1:1. The bone mineral density (BMD) values derived from DXA were applied as the reference standard. A three-class CNN model was developed to classify the patients into normal BMD, osteopenia, and osteoporosis. Moreover, we developed the models integrating the images with clinical covariates (age, gender, and BMI), and explored whether adding clinical data improves diagnostic performance over the image mode alone. The receiver operating characteristic curve analysis was performed for assessing the model performance.ResultsAs for classifying osteoporosis, the model based on the anteroposterior+lateral channel performed best, with the area under the curve (AUC) range from 0.909 to 0.937 in three test cohorts. The models with images alone achieved moderate sensitivity in classifying osteopenia, in which the highest AUC achieved 0.785. The performance of models integrating images with clinical data shows a slight improvement over models with anteroposterior or lateral images input alone for diagnosing osteoporosis, in which the AUC increased about 2%–4%. Regarding categorizing osteopenia and the normal BMD, the proposed models integrating images with clinical data also outperformed the models with images solely.ConclusionThe deep learning-based approach could screen osteoporosis and osteopenia based on lumbar radiographs. |
| first_indexed | 2024-04-12T19:21:47Z |
| format | Article |
| id | doaj.art-a13f5547a34d49a4b68b318b9353b1fe |
| institution | Directory Open Access Journal |
| issn | 1664-2392 |
| language | English |
| last_indexed | 2024-04-12T19:21:47Z |
| publishDate | 2022-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Endocrinology |
| spelling | doaj.art-a13f5547a34d49a4b68b318b9353b1fe2022-12-22T03:19:35ZengFrontiers Media S.A.Frontiers in Endocrinology1664-23922022-09-011310.3389/fendo.2022.971877971877Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese populationLiting Mao0Ziqiang Xia1Liang Pan2Jun Chen3Xian Liu4Zhiqiang Li5Zhaoxian Yan6Gengbin Lin7Huisen Wen8Bo Liu9Department of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of AI Research Lab, Guangzhou YLZ Ruitu Information Technology Co, Ltd, Guangzhou, ChinaDepartment of Radiology, ZHUHAI Branch of Guangdong Hospital of Chinese Medicine, Zhuhai, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaDepartment of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, ChinaPurposeMany high-risk osteopenia and osteoporosis patients remain undiagnosed. We proposed to construct a convolutional neural network model for screening primary osteopenia and osteoporosis based on the lumbar radiographs, and to compare the diagnostic performance of the CNN model adding the clinical covariates with the image model alone.MethodsA total of 6,908 participants were collected for analysis, including postmenopausal women and men aged 50–95 years, who performed conventional lumbar x-ray examinations and dual-energy x-ray absorptiometry (DXA) examinations within 3 months. All participants were divided into a training set, a validation set, test set 1, and test set 2 at a ratio of 8:1:1:1. The bone mineral density (BMD) values derived from DXA were applied as the reference standard. A three-class CNN model was developed to classify the patients into normal BMD, osteopenia, and osteoporosis. Moreover, we developed the models integrating the images with clinical covariates (age, gender, and BMI), and explored whether adding clinical data improves diagnostic performance over the image mode alone. The receiver operating characteristic curve analysis was performed for assessing the model performance.ResultsAs for classifying osteoporosis, the model based on the anteroposterior+lateral channel performed best, with the area under the curve (AUC) range from 0.909 to 0.937 in three test cohorts. The models with images alone achieved moderate sensitivity in classifying osteopenia, in which the highest AUC achieved 0.785. The performance of models integrating images with clinical data shows a slight improvement over models with anteroposterior or lateral images input alone for diagnosing osteoporosis, in which the AUC increased about 2%–4%. Regarding categorizing osteopenia and the normal BMD, the proposed models integrating images with clinical data also outperformed the models with images solely.ConclusionThe deep learning-based approach could screen osteoporosis and osteopenia based on lumbar radiographs.https://www.frontiersin.org/articles/10.3389/fendo.2022.971877/fullosteoporosisconvolutional neural network (CNN)screeningdual-energy x-ray absorptiometry (DXA)lumbar spine x-rays |
| spellingShingle | Liting Mao Ziqiang Xia Liang Pan Jun Chen Xian Liu Zhiqiang Li Zhaoxian Yan Gengbin Lin Huisen Wen Bo Liu Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population Frontiers in Endocrinology osteoporosis convolutional neural network (CNN) screening dual-energy x-ray absorptiometry (DXA) lumbar spine x-rays |
| title | Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population |
| title_full | Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population |
| title_fullStr | Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population |
| title_full_unstemmed | Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population |
| title_short | Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population |
| title_sort | deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a chinese population |
| topic | osteoporosis convolutional neural network (CNN) screening dual-energy x-ray absorptiometry (DXA) lumbar spine x-rays |
| url | https://www.frontiersin.org/articles/10.3389/fendo.2022.971877/full |
| work_keys_str_mv | AT litingmao deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT ziqiangxia deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT liangpan deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT junchen deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT xianliu deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT zhiqiangli deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT zhaoxianyan deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT gengbinlin deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT huisenwen deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation AT boliu deeplearningforscreeningprimaryosteopeniaandosteoporosisusingspineradiographsandpatientclinicalcovariatesinachinesepopulation |