Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN
Gastrointestinal endoscopy is widely conducted for the early detection of gastric cancer. However, it is often difficult to detect early gastric cancer lesions and accurately evaluate the invasive regions. Our study aimed to develop a detection and segmentation method for early gastric cancer region...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/11/3842 |
| _version_ | 1797566477125025792 |
|---|---|
| author | Tomoyuki Shibata Atsushi Teramoto Hyuga Yamada Naoki Ohmiya Kuniaki Saito Hiroshi Fujita |
| author_facet | Tomoyuki Shibata Atsushi Teramoto Hyuga Yamada Naoki Ohmiya Kuniaki Saito Hiroshi Fujita |
| author_sort | Tomoyuki Shibata |
| collection | DOAJ |
| description | Gastrointestinal endoscopy is widely conducted for the early detection of gastric cancer. However, it is often difficult to detect early gastric cancer lesions and accurately evaluate the invasive regions. Our study aimed to develop a detection and segmentation method for early gastric cancer regions from gastrointestinal endoscopic images. In this method, we first collected 1208 healthy and 533 cancer images. The gastric cancer region was detected and segmented from endoscopic images using Mask R-CNN, an instance segmentation method. An endoscopic image was provided to the Mask R-CNN, and a bounding box and a label image of the gastric cancer region were obtained. As a performance evaluation via five-fold cross-validation, sensitivity and false positives (FPs) per image were 96.0% and 0.10 FP/image, respectively. In the evaluation of segmentation of the gastric cancer region, the average Dice index was 71%. These results indicate that our proposed scheme may be useful for the detection of gastric cancer and evaluation of the invasive region in gastrointestinal endoscopy. |
| first_indexed | 2024-03-10T19:27:22Z |
| format | Article |
| id | doaj.art-0f9aa1962bcd4eb1b90a4dff764ac0bc |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T19:27:22Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-0f9aa1962bcd4eb1b90a4dff764ac0bc2023-11-20T02:23:31ZengMDPI AGApplied Sciences2076-34172020-05-011011384210.3390/app10113842Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNNTomoyuki Shibata0Atsushi Teramoto1Hyuga Yamada2Naoki Ohmiya3Kuniaki Saito4Hiroshi Fujita5School of Medicine, Fujita Health University, Toyoake, Aichi 470-1192, JapanGraduate School of Health Sciences, Fujita Health University, Toyoake, Aichi 470-1192, JapanSchool of Medicine, Fujita Health University, Toyoake, Aichi 470-1192, JapanSchool of Medicine, Fujita Health University, Toyoake, Aichi 470-1192, JapanGraduate School of Health Sciences, Fujita Health University, Toyoake, Aichi 470-1192, JapanFaculty of Engineering, Gifu University, Gifu, Gifu 501-1194, JapanGastrointestinal endoscopy is widely conducted for the early detection of gastric cancer. However, it is often difficult to detect early gastric cancer lesions and accurately evaluate the invasive regions. Our study aimed to develop a detection and segmentation method for early gastric cancer regions from gastrointestinal endoscopic images. In this method, we first collected 1208 healthy and 533 cancer images. The gastric cancer region was detected and segmented from endoscopic images using Mask R-CNN, an instance segmentation method. An endoscopic image was provided to the Mask R-CNN, and a bounding box and a label image of the gastric cancer region were obtained. As a performance evaluation via five-fold cross-validation, sensitivity and false positives (FPs) per image were 96.0% and 0.10 FP/image, respectively. In the evaluation of segmentation of the gastric cancer region, the average Dice index was 71%. These results indicate that our proposed scheme may be useful for the detection of gastric cancer and evaluation of the invasive region in gastrointestinal endoscopy.https://www.mdpi.com/2076-3417/10/11/3842deep learninggastric cancerendoscopysegmentationmask R-CNN |
| spellingShingle | Tomoyuki Shibata Atsushi Teramoto Hyuga Yamada Naoki Ohmiya Kuniaki Saito Hiroshi Fujita Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN Applied Sciences deep learning gastric cancer endoscopy segmentation mask R-CNN |
| title | Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN |
| title_full | Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN |
| title_fullStr | Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN |
| title_full_unstemmed | Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN |
| title_short | Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN |
| title_sort | automated detection and segmentation of early gastric cancer from endoscopic images using mask r cnn |
| topic | deep learning gastric cancer endoscopy segmentation mask R-CNN |
| url | https://www.mdpi.com/2076-3417/10/11/3842 |
| work_keys_str_mv | AT tomoyukishibata automateddetectionandsegmentationofearlygastriccancerfromendoscopicimagesusingmaskrcnn AT atsushiteramoto automateddetectionandsegmentationofearlygastriccancerfromendoscopicimagesusingmaskrcnn AT hyugayamada automateddetectionandsegmentationofearlygastriccancerfromendoscopicimagesusingmaskrcnn AT naokiohmiya automateddetectionandsegmentationofearlygastriccancerfromendoscopicimagesusingmaskrcnn AT kuniakisaito automateddetectionandsegmentationofearlygastriccancerfromendoscopicimagesusingmaskrcnn AT hiroshifujita automateddetectionandsegmentationofearlygastriccancerfromendoscopicimagesusingmaskrcnn |