Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images.
Deep learning technology has been used in the medical field to produce devices for clinical practice. Deep learning methods in cytology offer the potential to enhance cancer screening while also providing quantitative, objective, and highly reproducible testing. However, constructing high-accuracy d...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0285996 |
| _version_ | 1797801962535649280 |
|---|---|
| author | Yuki Kurita Shiori Meguro Naoko Tsuyama Isao Kosugi Yasunori Enomoto Hideya Kawasaki Takashi Uemura Michio Kimura Toshihide Iwashita |
| author_facet | Yuki Kurita Shiori Meguro Naoko Tsuyama Isao Kosugi Yasunori Enomoto Hideya Kawasaki Takashi Uemura Michio Kimura Toshihide Iwashita |
| author_sort | Yuki Kurita |
| collection | DOAJ |
| description | Deep learning technology has been used in the medical field to produce devices for clinical practice. Deep learning methods in cytology offer the potential to enhance cancer screening while also providing quantitative, objective, and highly reproducible testing. However, constructing high-accuracy deep learning models necessitates a significant amount of manually labeled data, which takes time. To address this issue, we used the Noisy Student Training technique to create a binary classification deep learning model for cervical cytology screening, which reduces the quantity of labeled data necessary. We used 140 whole-slide images from liquid-based cytology specimens, 50 of which were low-grade squamous intraepithelial lesions, 50 were high-grade squamous intraepithelial lesions, and 40 were negative samples. We extracted 56,996 images from the slides and then used them to train and test the model. We trained the EfficientNet using 2,600 manually labeled images to generate additional pseudo labels for the unlabeled data and then self-trained it within a student-teacher framework. Based on the presence or absence of abnormal cells, the created model was used to classify the images as normal or abnormal. The Grad-CAM approach was used to visualize the image components that contributed to the classification. The model achieved an area under the curve of 0.908, accuracy of 0.873, and F1-score of 0.833 with our test data. We also explored the optimal confidence threshold score and optimal augmentation approaches for low-magnification images. Our model efficiently classified normal and abnormal images at low magnification with high reliability, making it a promising screening tool for cervical cytology. |
| first_indexed | 2024-03-13T04:58:32Z |
| format | Article |
| id | doaj.art-3f8d4c6456874d3288487eb1e9ff2ef1 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-03-13T04:58:32Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-3f8d4c6456874d3288487eb1e9ff2ef12023-06-17T05:32:07ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01185e028599610.1371/journal.pone.0285996Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images.Yuki KuritaShiori MeguroNaoko TsuyamaIsao KosugiYasunori EnomotoHideya KawasakiTakashi UemuraMichio KimuraToshihide IwashitaDeep learning technology has been used in the medical field to produce devices for clinical practice. Deep learning methods in cytology offer the potential to enhance cancer screening while also providing quantitative, objective, and highly reproducible testing. However, constructing high-accuracy deep learning models necessitates a significant amount of manually labeled data, which takes time. To address this issue, we used the Noisy Student Training technique to create a binary classification deep learning model for cervical cytology screening, which reduces the quantity of labeled data necessary. We used 140 whole-slide images from liquid-based cytology specimens, 50 of which were low-grade squamous intraepithelial lesions, 50 were high-grade squamous intraepithelial lesions, and 40 were negative samples. We extracted 56,996 images from the slides and then used them to train and test the model. We trained the EfficientNet using 2,600 manually labeled images to generate additional pseudo labels for the unlabeled data and then self-trained it within a student-teacher framework. Based on the presence or absence of abnormal cells, the created model was used to classify the images as normal or abnormal. The Grad-CAM approach was used to visualize the image components that contributed to the classification. The model achieved an area under the curve of 0.908, accuracy of 0.873, and F1-score of 0.833 with our test data. We also explored the optimal confidence threshold score and optimal augmentation approaches for low-magnification images. Our model efficiently classified normal and abnormal images at low magnification with high reliability, making it a promising screening tool for cervical cytology.https://doi.org/10.1371/journal.pone.0285996 |
| spellingShingle | Yuki Kurita Shiori Meguro Naoko Tsuyama Isao Kosugi Yasunori Enomoto Hideya Kawasaki Takashi Uemura Michio Kimura Toshihide Iwashita Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images. PLoS ONE |
| title | Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images. |
| title_full | Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images. |
| title_fullStr | Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images. |
| title_full_unstemmed | Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images. |
| title_short | Accurate deep learning model using semi-supervised learning and Noisy Student for cervical cancer screening in low magnification images. |
| title_sort | accurate deep learning model using semi supervised learning and noisy student for cervical cancer screening in low magnification images |
| url | https://doi.org/10.1371/journal.pone.0285996 |
| work_keys_str_mv | AT yukikurita accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT shiorimeguro accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT naokotsuyama accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT isaokosugi accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT yasunorienomoto accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT hideyakawasaki accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT takashiuemura accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT michiokimura accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages AT toshihideiwashita accuratedeeplearningmodelusingsemisupervisedlearningandnoisystudentforcervicalcancerscreeninginlowmagnificationimages |