An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny
Automatic detection of Wireless Endoscopic Images can avoid dangerous possible diseases such as cancers. Therefore, a number of articles have been published on different methods to enhance the speed of detection and accuracy. We also present a custom version of the YOLOv4-tiny for Wireless Endoscopi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/21/10856 |
| _version_ | 1797469310522753024 |
|---|---|
| author | Mukhtorov Doniyorjon Rakhmonova Madinakhon Muksimova Shakhnoza Young-Im Cho |
| author_facet | Mukhtorov Doniyorjon Rakhmonova Madinakhon Muksimova Shakhnoza Young-Im Cho |
| author_sort | Mukhtorov Doniyorjon |
| collection | DOAJ |
| description | Automatic detection of Wireless Endoscopic Images can avoid dangerous possible diseases such as cancers. Therefore, a number of articles have been published on different methods to enhance the speed of detection and accuracy. We also present a custom version of the YOLOv4-tiny for Wireless Endoscopic Image detection and localization that uses a You Only Look Once (YOLO) version to enhance the model accuracy. We modified the YOLOv4-tiny model by replacing the CSPDarknet-53-tiny backbone structure with the Inception-ResNet-A block to enhance the accuracy of the original YOLOv4-tiny. In addition, we implemented a new custom data augmentation method to enhance the data quality, even for small datasets. We focused on maintaining the color of medical images because the sensitivity of medical images can affect the efficiency of the model. Experimental results showed that our proposed method obtains 99.4% training accuracy; compared with the previous models, this is more than a 1.2% increase. An original model used for both detection and the segmentation of medical images may cause a high error rate. In contrast, our proposed model could eliminate the error rate of the detection and localization of disease areas from wireless endoscopic images. |
| first_indexed | 2024-03-09T19:18:37Z |
| format | Article |
| id | doaj.art-f6e6cd648bb7497e8349347da97d7368 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T19:18:37Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-f6e6cd648bb7497e8349347da97d73682023-11-24T03:33:55ZengMDPI AGApplied Sciences2076-34172022-10-0112211085610.3390/app122110856An Improved Method of Polyp Detection Using Custom YOLOv4-TinyMukhtorov Doniyorjon0Rakhmonova Madinakhon1Muksimova Shakhnoza2Young-Im Cho3Department of IT Convergence Engineering, Gachon University, Sujeong-gu, Seongnam-si 461-701, KoreaDepartment of IT Convergence Engineering, Gachon University, Sujeong-gu, Seongnam-si 461-701, KoreaDepartment of IT Convergence Engineering, Gachon University, Sujeong-gu, Seongnam-si 461-701, KoreaDepartment of IT Convergence Engineering, Gachon University, Sujeong-gu, Seongnam-si 461-701, KoreaAutomatic detection of Wireless Endoscopic Images can avoid dangerous possible diseases such as cancers. Therefore, a number of articles have been published on different methods to enhance the speed of detection and accuracy. We also present a custom version of the YOLOv4-tiny for Wireless Endoscopic Image detection and localization that uses a You Only Look Once (YOLO) version to enhance the model accuracy. We modified the YOLOv4-tiny model by replacing the CSPDarknet-53-tiny backbone structure with the Inception-ResNet-A block to enhance the accuracy of the original YOLOv4-tiny. In addition, we implemented a new custom data augmentation method to enhance the data quality, even for small datasets. We focused on maintaining the color of medical images because the sensitivity of medical images can affect the efficiency of the model. Experimental results showed that our proposed method obtains 99.4% training accuracy; compared with the previous models, this is more than a 1.2% increase. An original model used for both detection and the segmentation of medical images may cause a high error rate. In contrast, our proposed model could eliminate the error rate of the detection and localization of disease areas from wireless endoscopic images.https://www.mdpi.com/2076-3417/12/21/10856wireless capsule endoscopy (WCE)polypclassificationneural networkslocalization |
| spellingShingle | Mukhtorov Doniyorjon Rakhmonova Madinakhon Muksimova Shakhnoza Young-Im Cho An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny Applied Sciences wireless capsule endoscopy (WCE) polyp classification neural networks localization |
| title | An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny |
| title_full | An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny |
| title_fullStr | An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny |
| title_full_unstemmed | An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny |
| title_short | An Improved Method of Polyp Detection Using Custom YOLOv4-Tiny |
| title_sort | improved method of polyp detection using custom yolov4 tiny |
| topic | wireless capsule endoscopy (WCE) polyp classification neural networks localization |
| url | https://www.mdpi.com/2076-3417/12/21/10856 |
| work_keys_str_mv | AT mukhtorovdoniyorjon animprovedmethodofpolypdetectionusingcustomyolov4tiny AT rakhmonovamadinakhon animprovedmethodofpolypdetectionusingcustomyolov4tiny AT muksimovashakhnoza animprovedmethodofpolypdetectionusingcustomyolov4tiny AT youngimcho animprovedmethodofpolypdetectionusingcustomyolov4tiny AT mukhtorovdoniyorjon improvedmethodofpolypdetectionusingcustomyolov4tiny AT rakhmonovamadinakhon improvedmethodofpolypdetectionusingcustomyolov4tiny AT muksimovashakhnoza improvedmethodofpolypdetectionusingcustomyolov4tiny AT youngimcho improvedmethodofpolypdetectionusingcustomyolov4tiny |