Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images
PurposeTo develop and validate a deep learning radiomics (DLR) model that uses X-ray images to predict the classification of osteoporotic vertebral fractures (OVFs).Material and methodsThe study encompassed a cohort of 942 patients, involving examinations of 1076 vertebrae through X-ray, CT, and MRI...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-03-01
|
| Series: | Frontiers in Endocrinology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fendo.2024.1370838/full |
| _version_ | 1827304395256102912 |
|---|---|
| author | Jun Zhang Jun Zhang Liang Xia Jiayi Liu Xiaoying Niu Jun Tang Jianguo Xia Yongkang Liu Weixiao Zhang Zhipeng Liang Xueli Zhang Guangyu Tang Guangyu Tang Lin Zhang |
| author_facet | Jun Zhang Jun Zhang Liang Xia Jiayi Liu Xiaoying Niu Jun Tang Jianguo Xia Yongkang Liu Weixiao Zhang Zhipeng Liang Xueli Zhang Guangyu Tang Guangyu Tang Lin Zhang |
| author_sort | Jun Zhang |
| collection | DOAJ |
| description | PurposeTo develop and validate a deep learning radiomics (DLR) model that uses X-ray images to predict the classification of osteoporotic vertebral fractures (OVFs).Material and methodsThe study encompassed a cohort of 942 patients, involving examinations of 1076 vertebrae through X-ray, CT, and MRI across three distinct hospitals. The OVFs were categorized as class 0, 1, or 2 based on the Assessment System of Thoracolumbar Osteoporotic Fracture. The dataset was divided randomly into four distinct subsets: a training set comprising 712 samples, an internal validation set with 178 samples, an external validation set containing 111 samples, and a prospective validation set consisting of 75 samples. The ResNet-50 architectural model was used to implement deep transfer learning (DTL), undergoing -pre-training separately on the RadImageNet and ImageNet datasets. Features from DTL and radiomics were extracted and integrated using X-ray images. The optimal fusion feature model was identified through least absolute shrinkage and selection operator logistic regression. Evaluation of the predictive capabilities for OVFs classification involved eight machine learning models, assessed through receiver operating characteristic curves employing the “One-vs-Rest” strategy. The Delong test was applied to compare the predictive performance of the superior RadImageNet model against the ImageNet model.ResultsFollowing pre-training separately on RadImageNet and ImageNet datasets, feature selection and fusion yielded 17 and 12 fusion features, respectively. Logistic regression emerged as the optimal machine learning algorithm for both DLR models. Across the training set, internal validation set, external validation set, and prospective validation set, the macro-average Area Under the Curve (AUC) based on the RadImageNet dataset surpassed those based on the ImageNet dataset, with statistically significant differences observed (P<0.05). Utilizing the binary “One-vs-Rest” strategy, the model based on the RadImageNet dataset demonstrated superior efficacy in predicting Class 0, achieving an AUC of 0.969 and accuracy of 0.863. Predicting Class 1 yielded an AUC of 0.945 and accuracy of 0.875, while for Class 2, the AUC and accuracy were 0.809 and 0.692, respectively.ConclusionThe DLR model, based on the RadImageNet dataset, outperformed the ImageNet model in predicting the classification of OVFs, with generalizability confirmed in the prospective validation set. |
| first_indexed | 2024-04-24T17:36:00Z |
| format | Article |
| id | doaj.art-537f166ae4d04230828e708f75a08e7e |
| institution | Directory Open Access Journal |
| issn | 1664-2392 |
| language | English |
| last_indexed | 2024-04-24T17:36:00Z |
| publishDate | 2024-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Endocrinology |
| spelling | doaj.art-537f166ae4d04230828e708f75a08e7e2024-03-28T05:00:24ZengFrontiers Media S.A.Frontiers in Endocrinology1664-23922024-03-011510.3389/fendo.2024.13708381370838Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray imagesJun Zhang0Jun Zhang1Liang Xia2Jiayi Liu3Xiaoying Niu4Jun Tang5Jianguo Xia6Yongkang Liu7Weixiao Zhang8Zhipeng Liang9Xueli Zhang10Guangyu Tang11Guangyu Tang12Lin Zhang13Department of Radiology, Shanghai Tenth People’s Hospital, Clinical Medical College of Nanjing Medical University, Shanghai, ChinaDepartment of Radiology, Sir RunRun Hospital, Nanjing Medical University, Nanjing, ChinaDepartment of Radiology, Sir RunRun Hospital, Nanjing Medical University, Nanjing, ChinaDepartment of Radiology, Sir RunRun Hospital, Nanjing Medical University, Nanjing, ChinaDepartment of Neonates, Dongfeng General Hospital of National Medicine, Hubei University of Medicine, Shiyan, ChinaDepartment of Radiology, The Affiliated Taizhou People’s Hospital of Nanjing Medical University, Taizhou, ChinaDepartment of Radiology, The Affiliated Taizhou People’s Hospital of Nanjing Medical University, Taizhou, ChinaDepartment of Radiology, Jiangsu Provincial Hospital of Traditional Chinese Medicine, Nanjing University of Chinese Medicine, Nanjing, ChinaDepartment of Radiology, Sir RunRun Hospital, Nanjing Medical University, Nanjing, ChinaDepartment of Radiology, Sir RunRun Hospital, Nanjing Medical University, Nanjing, ChinaDepartment of Radiology, Shanghai Tenth People’s Hospital, Tongji University School of Medicine, Shanghai, ChinaDepartment of Radiology, Shanghai Tenth People’s Hospital, Clinical Medical College of Nanjing Medical University, Shanghai, ChinaDepartment of Radiology, Shanghai Tenth People’s Hospital, Tongji University School of Medicine, Shanghai, ChinaDepartment of Radiology, Shanghai Tenth People’s Hospital, Tongji University School of Medicine, Shanghai, ChinaPurposeTo develop and validate a deep learning radiomics (DLR) model that uses X-ray images to predict the classification of osteoporotic vertebral fractures (OVFs).Material and methodsThe study encompassed a cohort of 942 patients, involving examinations of 1076 vertebrae through X-ray, CT, and MRI across three distinct hospitals. The OVFs were categorized as class 0, 1, or 2 based on the Assessment System of Thoracolumbar Osteoporotic Fracture. The dataset was divided randomly into four distinct subsets: a training set comprising 712 samples, an internal validation set with 178 samples, an external validation set containing 111 samples, and a prospective validation set consisting of 75 samples. The ResNet-50 architectural model was used to implement deep transfer learning (DTL), undergoing -pre-training separately on the RadImageNet and ImageNet datasets. Features from DTL and radiomics were extracted and integrated using X-ray images. The optimal fusion feature model was identified through least absolute shrinkage and selection operator logistic regression. Evaluation of the predictive capabilities for OVFs classification involved eight machine learning models, assessed through receiver operating characteristic curves employing the “One-vs-Rest” strategy. The Delong test was applied to compare the predictive performance of the superior RadImageNet model against the ImageNet model.ResultsFollowing pre-training separately on RadImageNet and ImageNet datasets, feature selection and fusion yielded 17 and 12 fusion features, respectively. Logistic regression emerged as the optimal machine learning algorithm for both DLR models. Across the training set, internal validation set, external validation set, and prospective validation set, the macro-average Area Under the Curve (AUC) based on the RadImageNet dataset surpassed those based on the ImageNet dataset, with statistically significant differences observed (P<0.05). Utilizing the binary “One-vs-Rest” strategy, the model based on the RadImageNet dataset demonstrated superior efficacy in predicting Class 0, achieving an AUC of 0.969 and accuracy of 0.863. Predicting Class 1 yielded an AUC of 0.945 and accuracy of 0.875, while for Class 2, the AUC and accuracy were 0.809 and 0.692, respectively.ConclusionThe DLR model, based on the RadImageNet dataset, outperformed the ImageNet model in predicting the classification of OVFs, with generalizability confirmed in the prospective validation set.https://www.frontiersin.org/articles/10.3389/fendo.2024.1370838/fullosteoporotic vertebral fracturesclassificationX-ray computed tomographydeep learningradiomics |
| spellingShingle | Jun Zhang Jun Zhang Liang Xia Jiayi Liu Xiaoying Niu Jun Tang Jianguo Xia Yongkang Liu Weixiao Zhang Zhipeng Liang Xueli Zhang Guangyu Tang Guangyu Tang Lin Zhang Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images Frontiers in Endocrinology osteoporotic vertebral fractures classification X-ray computed tomography deep learning radiomics |
| title | Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images |
| title_full | Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images |
| title_fullStr | Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images |
| title_full_unstemmed | Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images |
| title_short | Exploring deep learning radiomics for classifying osteoporotic vertebral fractures in X-ray images |
| title_sort | exploring deep learning radiomics for classifying osteoporotic vertebral fractures in x ray images |
| topic | osteoporotic vertebral fractures classification X-ray computed tomography deep learning radiomics |
| url | https://www.frontiersin.org/articles/10.3389/fendo.2024.1370838/full |
| work_keys_str_mv | AT junzhang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT junzhang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT liangxia exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT jiayiliu exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT xiaoyingniu exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT juntang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT jianguoxia exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT yongkangliu exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT weixiaozhang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT zhipengliang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT xuelizhang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT guangyutang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT guangyutang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages AT linzhang exploringdeeplearningradiomicsforclassifyingosteoporoticvertebralfracturesinxrayimages |