Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation
In today’s high-order health examination, imaging examination accounts for a large proportion. Computed tomography (CT), which can detect the whole body, uses X-rays to penetrate the human body to obtain images. Its presentation is a high-resolution black-and-white image composed of gray scales. It...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/15/4/1343 |
| _version_ | 1797621869141032960 |
|---|---|
| author | Ming-Chan Lee Shao-Yu Wang Cheng-Tang Pan Ming-Yi Chien Wei-Ming Li Jin-Hao Xu Chi-Hung Luo Yow-Ling Shiue |
| author_facet | Ming-Chan Lee Shao-Yu Wang Cheng-Tang Pan Ming-Yi Chien Wei-Ming Li Jin-Hao Xu Chi-Hung Luo Yow-Ling Shiue |
| author_sort | Ming-Chan Lee |
| collection | DOAJ |
| description | In today’s high-order health examination, imaging examination accounts for a large proportion. Computed tomography (CT), which can detect the whole body, uses X-rays to penetrate the human body to obtain images. Its presentation is a high-resolution black-and-white image composed of gray scales. It is expected to assist doctors in making judgments through deep learning based on the image recognition technology of artificial intelligence. It used CT images to identify the bladder and lesions and then segmented them in the images. The images can achieve high accuracy without using a developer. In this study, the U-Net neural network, commonly used in the medical field, was used to extend the encoder position in combination with the ResBlock in ResNet and the Dense Block in DenseNet, so that the training could maintain the training parameters while reducing the overall identification operation time. The decoder could be used in combination with Attention Gates to suppress the irrelevant areas of the image while paying attention to significant features. Combined with the above algorithm, we proposed a Residual-Dense Attention (RDA) U-Net model, which was used to identify organs and lesions from CT images of abdominal scans. The accuracy (<i>ACC</i>) of using this model for the bladder and its lesions was 96% and 93%, respectively. The values of Intersection over Union (<i>IoU</i>) were 0.9505 and 0.8024, respectively. Average Hausdorff distance (<i>AVGDIST</i>) was as low as 0.02 and 0.12, respectively, and the overall training time was reduced by up to 44% compared with other convolution neural networks. |
| first_indexed | 2024-03-11T09:02:09Z |
| format | Article |
| id | doaj.art-9a660155a0944f41aac5ce02482c45c3 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-11T09:02:09Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-9a660155a0944f41aac5ce02482c45c32023-11-16T19:39:42ZengMDPI AGCancers2072-66942023-02-01154134310.3390/cancers15041343Development of Deep Learning with RDA U-Net Network for Bladder Cancer SegmentationMing-Chan Lee0Shao-Yu Wang1Cheng-Tang Pan2Ming-Yi Chien3Wei-Ming Li4Jin-Hao Xu5Chi-Hung Luo6Yow-Ling Shiue7Department of Mechanical and Electro-Mechanical Engineering, National Sun Yat-sen University, Kaohsiung 804, TaiwanDepartment of Mechanical Engineering, National United University, Miaoli 360, TaiwanDepartment of Mechanical and Electro-Mechanical Engineering, National Sun Yat-sen University, Kaohsiung 804, TaiwanDepartment of Mechanical and Electro-Mechanical Engineering, National Sun Yat-sen University, Kaohsiung 804, TaiwanDepartment of Urology, Kaohsiung Medical University, Kaohsiung 807, TaiwanDepartment of Medicine Chest, Kaohsiung Armed Forces General Hospital, Kaohsiung 802, TaiwanDepartment of Medicine Chest, Kaohsiung Armed Forces General Hospital, Kaohsiung 802, TaiwanInstitute of Biomedical Sciences, National Sun Yat-sen University, Kaohsiung 804, TaiwanIn today’s high-order health examination, imaging examination accounts for a large proportion. Computed tomography (CT), which can detect the whole body, uses X-rays to penetrate the human body to obtain images. Its presentation is a high-resolution black-and-white image composed of gray scales. It is expected to assist doctors in making judgments through deep learning based on the image recognition technology of artificial intelligence. It used CT images to identify the bladder and lesions and then segmented them in the images. The images can achieve high accuracy without using a developer. In this study, the U-Net neural network, commonly used in the medical field, was used to extend the encoder position in combination with the ResBlock in ResNet and the Dense Block in DenseNet, so that the training could maintain the training parameters while reducing the overall identification operation time. The decoder could be used in combination with Attention Gates to suppress the irrelevant areas of the image while paying attention to significant features. Combined with the above algorithm, we proposed a Residual-Dense Attention (RDA) U-Net model, which was used to identify organs and lesions from CT images of abdominal scans. The accuracy (<i>ACC</i>) of using this model for the bladder and its lesions was 96% and 93%, respectively. The values of Intersection over Union (<i>IoU</i>) were 0.9505 and 0.8024, respectively. Average Hausdorff distance (<i>AVGDIST</i>) was as low as 0.02 and 0.12, respectively, and the overall training time was reduced by up to 44% compared with other convolution neural networks.https://www.mdpi.com/2072-6694/15/4/1343computed tomographyU-Netbladder cancerResNetDenseNet |
| spellingShingle | Ming-Chan Lee Shao-Yu Wang Cheng-Tang Pan Ming-Yi Chien Wei-Ming Li Jin-Hao Xu Chi-Hung Luo Yow-Ling Shiue Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation Cancers computed tomography U-Net bladder cancer ResNet DenseNet |
| title | Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation |
| title_full | Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation |
| title_fullStr | Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation |
| title_full_unstemmed | Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation |
| title_short | Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation |
| title_sort | development of deep learning with rda u net network for bladder cancer segmentation |
| topic | computed tomography U-Net bladder cancer ResNet DenseNet |
| url | https://www.mdpi.com/2072-6694/15/4/1343 |
| work_keys_str_mv | AT mingchanlee developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT shaoyuwang developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT chengtangpan developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT mingyichien developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT weimingli developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT jinhaoxu developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT chihungluo developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation AT yowlingshiue developmentofdeeplearningwithrdaunetnetworkforbladdercancersegmentation |