Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning
ObjectiveTo develop a deep learning-based model using esophageal thickness to detect esophageal cancer from unenhanced chest CT images.MethodsWe retrospectively identified 141 patients with esophageal cancer and 273 patients negative for esophageal cancer (at the time of imaging) for model training....
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-09-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2021.700210/full |
| _version_ | 1818682812864135168 |
|---|---|
| author | He Sui Ruhang Ma Ruhang Ma Lin Liu Yaozong Gao Wenhai Zhang Zhanhao Mo |
| author_facet | He Sui Ruhang Ma Ruhang Ma Lin Liu Yaozong Gao Wenhai Zhang Zhanhao Mo |
| author_sort | He Sui |
| collection | DOAJ |
| description | ObjectiveTo develop a deep learning-based model using esophageal thickness to detect esophageal cancer from unenhanced chest CT images.MethodsWe retrospectively identified 141 patients with esophageal cancer and 273 patients negative for esophageal cancer (at the time of imaging) for model training. Unenhanced chest CT images were collected and used to build a convolutional neural network (CNN) model for diagnosing esophageal cancer. The CNN is a VB-Net segmentation network that segments the esophagus and automatically quantifies the thickness of the esophageal wall and detect positions of esophageal lesions. To validate this model, 52 false negatives and 48 normal cases were collected further as the second dataset. The average performance of three radiologists and that of the same radiologists aided by the model were compared.ResultsThe sensitivity and specificity of the esophageal cancer detection model were 88.8% and 90.9%, respectively, for the validation dataset set. Of the 52 missed esophageal cancer cases and the 48 normal cases, the sensitivity, specificity, and accuracy of the deep learning esophageal cancer detection model were 69%, 61%, and 65%, respectively. The independent results of the radiologists had a sensitivity of 25%, 31%, and 27%; specificity of 78%, 75%, and 75%; and accuracy of 53%, 54%, and 53%. With the aid of the model, the results of the radiologists were improved to a sensitivity of 77%, 81%, and 75%; specificity of 75%, 74%, and 74%; and accuracy of 76%, 77%, and 75%, respectively.ConclusionsDeep learning-based model can effectively detect esophageal cancer in unenhanced chest CT scans to improve the incidental detection of esophageal cancer. |
| first_indexed | 2024-12-17T10:24:48Z |
| format | Article |
| id | doaj.art-65c20e8c9a80445693a3991b43fde51a |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-12-17T10:24:48Z |
| publishDate | 2021-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-65c20e8c9a80445693a3991b43fde51a2022-12-21T21:52:41ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2021-09-011110.3389/fonc.2021.700210700210Detection of Incidental Esophageal Cancers on Chest CT by Deep LearningHe Sui0Ruhang Ma1Ruhang Ma2Lin Liu3Yaozong Gao4Wenhai Zhang5Zhanhao Mo6China-Japan Union Hospital of Jilin University, Changchun, ChinaChina-Japan Union Hospital of Jilin University, Changchun, ChinaRadiology Department, Weifang People’s Hospital, Weifang, ChinaChina-Japan Union Hospital of Jilin University, Changchun, ChinaShanghai United Imaging Medical Technology Co., Ltd., Shanghai, ChinaShanghai United Imaging Medical Technology Co., Ltd., Shanghai, ChinaChina-Japan Union Hospital of Jilin University, Changchun, ChinaObjectiveTo develop a deep learning-based model using esophageal thickness to detect esophageal cancer from unenhanced chest CT images.MethodsWe retrospectively identified 141 patients with esophageal cancer and 273 patients negative for esophageal cancer (at the time of imaging) for model training. Unenhanced chest CT images were collected and used to build a convolutional neural network (CNN) model for diagnosing esophageal cancer. The CNN is a VB-Net segmentation network that segments the esophagus and automatically quantifies the thickness of the esophageal wall and detect positions of esophageal lesions. To validate this model, 52 false negatives and 48 normal cases were collected further as the second dataset. The average performance of three radiologists and that of the same radiologists aided by the model were compared.ResultsThe sensitivity and specificity of the esophageal cancer detection model were 88.8% and 90.9%, respectively, for the validation dataset set. Of the 52 missed esophageal cancer cases and the 48 normal cases, the sensitivity, specificity, and accuracy of the deep learning esophageal cancer detection model were 69%, 61%, and 65%, respectively. The independent results of the radiologists had a sensitivity of 25%, 31%, and 27%; specificity of 78%, 75%, and 75%; and accuracy of 53%, 54%, and 53%. With the aid of the model, the results of the radiologists were improved to a sensitivity of 77%, 81%, and 75%; specificity of 75%, 74%, and 74%; and accuracy of 76%, 77%, and 75%, respectively.ConclusionsDeep learning-based model can effectively detect esophageal cancer in unenhanced chest CT scans to improve the incidental detection of esophageal cancer.https://www.frontiersin.org/articles/10.3389/fonc.2021.700210/fulldeep learningconvolutional neural networkchest CTesophageal cancerv-net |
| spellingShingle | He Sui Ruhang Ma Ruhang Ma Lin Liu Yaozong Gao Wenhai Zhang Zhanhao Mo Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning Frontiers in Oncology deep learning convolutional neural network chest CT esophageal cancer v-net |
| title | Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning |
| title_full | Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning |
| title_fullStr | Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning |
| title_full_unstemmed | Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning |
| title_short | Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning |
| title_sort | detection of incidental esophageal cancers on chest ct by deep learning |
| topic | deep learning convolutional neural network chest CT esophageal cancer v-net |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2021.700210/full |
| work_keys_str_mv | AT hesui detectionofincidentalesophagealcancersonchestctbydeeplearning AT ruhangma detectionofincidentalesophagealcancersonchestctbydeeplearning AT ruhangma detectionofincidentalesophagealcancersonchestctbydeeplearning AT linliu detectionofincidentalesophagealcancersonchestctbydeeplearning AT yaozonggao detectionofincidentalesophagealcancersonchestctbydeeplearning AT wenhaizhang detectionofincidentalesophagealcancersonchestctbydeeplearning AT zhanhaomo detectionofincidentalesophagealcancersonchestctbydeeplearning |