Automatic Classification of Cervical Cells Using Deep Learning Method
Cervical cancer is the fourth most prevalent disease among women. Prompt diagnosis and its management can significantly improve patients' survival rates. Therefore, routine screening for cervical cancer is of paramount importance. Herein, we explore the potential of a deep learning model to aut...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9358146/ |
| _version_ | 1819180745758867456 |
|---|---|
| author | Suxiang Yu Xinxing Feng Bin Wang Hua Dun Shuai Zhang Ruihong Zhang Xin Huang |
| author_facet | Suxiang Yu Xinxing Feng Bin Wang Hua Dun Shuai Zhang Ruihong Zhang Xin Huang |
| author_sort | Suxiang Yu |
| collection | DOAJ |
| description | Cervical cancer is the fourth most prevalent disease among women. Prompt diagnosis and its management can significantly improve patients' survival rates. Therefore, routine screening for cervical cancer is of paramount importance. Herein, we explore the potential of a deep learning model to automatically distinguish abnormal cells from normal cells. The ThinPrep cytologic test dataset was collected from the fourth central hospital of Baoding city, China. Based on the dataset, four classification models were developed. The first model was a 10-layer convolutional neural network (CNN). The second model was an advancement of the first model equipped with a spatial pyramid pooling (SPP) layer (CNN + SPP) to treat cell images based on their sizes. Based on the first model, the third model replaced the CNN layers with the inception module (CNN + Inception). However, the fourth model incorporated both the SPP layer and the inception module into the first model (CNN + inception + SPP). The performances of the four models are estimated and compared by using the same testing data and evaluation index. The testing results demonstrated that the fourth model yields the best performance. Moreover, the area under the curve (AUC) for module four was 0.997. |
| first_indexed | 2024-12-22T22:19:14Z |
| format | Article |
| id | doaj.art-6182239845bd4dc59816556d6284b508 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T22:19:14Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-6182239845bd4dc59816556d6284b5082022-12-21T18:10:42ZengIEEEIEEE Access2169-35362021-01-019325593256810.1109/ACCESS.2021.30604479358146Automatic Classification of Cervical Cells Using Deep Learning MethodSuxiang Yu0Xinxing Feng1https://orcid.org/0000-0002-4781-2116Bin Wang2Hua Dun3Shuai Zhang4https://orcid.org/0000-0002-5821-2650Ruihong Zhang5Xin Huang6https://orcid.org/0000-0003-0569-5932Department of Pathology, The Fourth Central Hospital of Baoding City, Baoding, ChinaEndocrinology and Cardiovascular Disease Centre, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, ChinaDepartment of Pathology, The Fourth Central Hospital of Baoding City, Baoding, ChinaDepartment of Pathology, The Fourth Central Hospital of Baoding City, Baoding, ChinaDepartment of Computer Science, The University of Manchester, Manchester, U.K.Department of Science and Teaching, The Fourth Central Hospital of Baoding City, Baoding, ChinaSolar Activity Prediction Center, National Astronomical Observatories, Chinese Academy of Sciences, Beijing, ChinaCervical cancer is the fourth most prevalent disease among women. Prompt diagnosis and its management can significantly improve patients' survival rates. Therefore, routine screening for cervical cancer is of paramount importance. Herein, we explore the potential of a deep learning model to automatically distinguish abnormal cells from normal cells. The ThinPrep cytologic test dataset was collected from the fourth central hospital of Baoding city, China. Based on the dataset, four classification models were developed. The first model was a 10-layer convolutional neural network (CNN). The second model was an advancement of the first model equipped with a spatial pyramid pooling (SPP) layer (CNN + SPP) to treat cell images based on their sizes. Based on the first model, the third model replaced the CNN layers with the inception module (CNN + Inception). However, the fourth model incorporated both the SPP layer and the inception module into the first model (CNN + inception + SPP). The performances of the four models are estimated and compared by using the same testing data and evaluation index. The testing results demonstrated that the fourth model yields the best performance. Moreover, the area under the curve (AUC) for module four was 0.997.https://ieeexplore.ieee.org/document/9358146/Cell classificationdeep learningneural networkscervical cytology |
| spellingShingle | Suxiang Yu Xinxing Feng Bin Wang Hua Dun Shuai Zhang Ruihong Zhang Xin Huang Automatic Classification of Cervical Cells Using Deep Learning Method IEEE Access Cell classification deep learning neural networks cervical cytology |
| title | Automatic Classification of Cervical Cells Using Deep Learning Method |
| title_full | Automatic Classification of Cervical Cells Using Deep Learning Method |
| title_fullStr | Automatic Classification of Cervical Cells Using Deep Learning Method |
| title_full_unstemmed | Automatic Classification of Cervical Cells Using Deep Learning Method |
| title_short | Automatic Classification of Cervical Cells Using Deep Learning Method |
| title_sort | automatic classification of cervical cells using deep learning method |
| topic | Cell classification deep learning neural networks cervical cytology |
| url | https://ieeexplore.ieee.org/document/9358146/ |
| work_keys_str_mv | AT suxiangyu automaticclassificationofcervicalcellsusingdeeplearningmethod AT xinxingfeng automaticclassificationofcervicalcellsusingdeeplearningmethod AT binwang automaticclassificationofcervicalcellsusingdeeplearningmethod AT huadun automaticclassificationofcervicalcellsusingdeeplearningmethod AT shuaizhang automaticclassificationofcervicalcellsusingdeeplearningmethod AT ruihongzhang automaticclassificationofcervicalcellsusingdeeplearningmethod AT xinhuang automaticclassificationofcervicalcellsusingdeeplearningmethod |