Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study
Ankle fractures are common and, compared to other injuries, tend to be overlooked in the emergency department. We aim to develop a deep learning algorithm that can detect not only definite fractures but also obscure fractures. We collected the data of 1226 patients with suspected ankle fractures and...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/11/19/8791 |
| _version_ | 1797516855806525440 |
|---|---|
| author | Ji-Hun Kim Yong-Cheol Mo Seung-Myung Choi Youk Hyun Jung Woo Lee |
| author_facet | Ji-Hun Kim Yong-Cheol Mo Seung-Myung Choi Youk Hyun Jung Woo Lee |
| author_sort | Ji-Hun Kim |
| collection | DOAJ |
| description | Ankle fractures are common and, compared to other injuries, tend to be overlooked in the emergency department. We aim to develop a deep learning algorithm that can detect not only definite fractures but also obscure fractures. We collected the data of 1226 patients with suspected ankle fractures and performed both X-rays and CT scans. With anteroposterior (AP) and lateral ankle X-rays of 1040 patients with fractures and 186 normal patients, we developed a deep learning model. The training, validation, and test datasets were split in a 3/1/1 ratio. Data augmentation and under-sampling techniques were administered as part of the preprocessing. The Inception V3 model was utilized for the image classification. Performance of the model was validated using a confusion matrix and the area under the receiver operating characteristic curve (AUC-ROC). For the AP and lateral trials, the best accuracy and AUC values were 83%/0.91 in AP and 90%/0.95 in lateral. Additionally, the mean accuracy and AUC values were 83%/0.89 for the AP trials and 83%/0.9 for the lateral trials. The reliable dataset resulted in the CNN model providing higher accuracy than in past studies. |
| first_indexed | 2024-03-10T07:07:41Z |
| format | Article |
| id | doaj.art-66c9903954ac47d380f82b5c8b838f30 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T07:07:41Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-66c9903954ac47d380f82b5c8b838f302023-11-22T15:43:03ZengMDPI AGApplied Sciences2076-34172021-09-011119879110.3390/app11198791Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational StudyJi-Hun Kim0Yong-Cheol Mo1Seung-Myung Choi2Youk Hyun3Jung Woo Lee4Department of Emergency Medicine, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Chuncheon 24252, KoreaDepartment of Mathematics, Sungkyunkwan University, Suwon 16419, KoreaDepartment of Orthopedic Surgery, Uijeongbu Eulji Medical Center, Eulji University, Daejeon 34824, KoreaDepartment of Emergency Medicine, Wonju College of Medicine, Yonsei University, Wonju 26426, KoreaBigdata Platform Business Group, Wonju Yonsei Medical Center, Yonsei University, Wonju 26426, KoreaAnkle fractures are common and, compared to other injuries, tend to be overlooked in the emergency department. We aim to develop a deep learning algorithm that can detect not only definite fractures but also obscure fractures. We collected the data of 1226 patients with suspected ankle fractures and performed both X-rays and CT scans. With anteroposterior (AP) and lateral ankle X-rays of 1040 patients with fractures and 186 normal patients, we developed a deep learning model. The training, validation, and test datasets were split in a 3/1/1 ratio. Data augmentation and under-sampling techniques were administered as part of the preprocessing. The Inception V3 model was utilized for the image classification. Performance of the model was validated using a confusion matrix and the area under the receiver operating characteristic curve (AUC-ROC). For the AP and lateral trials, the best accuracy and AUC values were 83%/0.91 in AP and 90%/0.95 in lateral. Additionally, the mean accuracy and AUC values were 83%/0.89 for the AP trials and 83%/0.9 for the lateral trials. The reliable dataset resulted in the CNN model providing higher accuracy than in past studies.https://www.mdpi.com/2076-3417/11/19/8791ankle fracturescomputed tomographydeep learningartificial intelligenceconvolutional neural network |
| spellingShingle | Ji-Hun Kim Yong-Cheol Mo Seung-Myung Choi Youk Hyun Jung Woo Lee Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study Applied Sciences ankle fractures computed tomography deep learning artificial intelligence convolutional neural network |
| title | Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study |
| title_full | Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study |
| title_fullStr | Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study |
| title_full_unstemmed | Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study |
| title_short | Detecting Ankle Fractures in Plain Radiographs Using Deep Learning with Accurately Labeled Datasets Aided by Computed Tomography: A Retrospective Observational Study |
| title_sort | detecting ankle fractures in plain radiographs using deep learning with accurately labeled datasets aided by computed tomography a retrospective observational study |
| topic | ankle fractures computed tomography deep learning artificial intelligence convolutional neural network |
| url | https://www.mdpi.com/2076-3417/11/19/8791 |
| work_keys_str_mv | AT jihunkim detectinganklefracturesinplainradiographsusingdeeplearningwithaccuratelylabeleddatasetsaidedbycomputedtomographyaretrospectiveobservationalstudy AT yongcheolmo detectinganklefracturesinplainradiographsusingdeeplearningwithaccuratelylabeleddatasetsaidedbycomputedtomographyaretrospectiveobservationalstudy AT seungmyungchoi detectinganklefracturesinplainradiographsusingdeeplearningwithaccuratelylabeleddatasetsaidedbycomputedtomographyaretrospectiveobservationalstudy AT youkhyun detectinganklefracturesinplainradiographsusingdeeplearningwithaccuratelylabeleddatasetsaidedbycomputedtomographyaretrospectiveobservationalstudy AT jungwoolee detectinganklefracturesinplainradiographsusingdeeplearningwithaccuratelylabeleddatasetsaidedbycomputedtomographyaretrospectiveobservationalstudy |