Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms
PurposeTo automatically evaluate renal masses in CT images by using a cascade 3D U-Net- and ResNet-based method to accurately segment and classify focal renal lesions.Material and MethodsWe used an institutional dataset comprising 610 CT image series from 490 patients from August 2009 to August 2021...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-05-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2023.1169922/full |
| _version_ | 1797824126842306560 |
|---|---|
| author | Tongtong Zhao Zhaonan Sun Ying Guo Yumeng Sun Yaofeng Zhang Xiaoying Wang |
| author_facet | Tongtong Zhao Zhaonan Sun Ying Guo Yumeng Sun Yaofeng Zhang Xiaoying Wang |
| author_sort | Tongtong Zhao |
| collection | DOAJ |
| description | PurposeTo automatically evaluate renal masses in CT images by using a cascade 3D U-Net- and ResNet-based method to accurately segment and classify focal renal lesions.Material and MethodsWe used an institutional dataset comprising 610 CT image series from 490 patients from August 2009 to August 2021 to train and evaluate the proposed method. We first determined the boundaries of the kidneys on the CT images utilizing a 3D U-Net-based method to be used as a region of interest to search for renal mass. An ensemble learning model based on 3D U-Net was then used to detect and segment the masses, followed by a ResNet algorithm for classification. Our algorithm was evaluated with an external validation dataset and kidney tumor segmentation (KiTS21) challenge dataset.ResultsThe algorithm achieved a Dice similarity coefficient (DSC) of 0.99 for bilateral kidney boundary segmentation in the test set. The average DSC for renal mass delineation using the 3D U-Net was 0.75 and 0.83. Our method detected renal masses with recalls of 84.54% and 75.90%. The classification accuracy in the test set was 86.05% for masses (<5 mm) and 91.97% for masses (≥5 mm).ConclusionWe developed a deep learning-based method for fully automated segmentation and classification of renal masses in CT images. Testing of this algorithm showed that it has the capability of accurately localizing and classifying renal masses. |
| first_indexed | 2024-03-13T10:34:22Z |
| format | Article |
| id | doaj.art-cbace3a0735449f6b9c3dab97cb5f866 |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-03-13T10:34:22Z |
| publishDate | 2023-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-cbace3a0735449f6b9c3dab97cb5f8662023-05-18T08:12:04ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2023-05-011310.3389/fonc.2023.11699221169922Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithmsTongtong Zhao0Zhaonan Sun1Ying Guo2Yumeng Sun3Yaofeng Zhang4Xiaoying Wang5Department of Radiology, Peking University First Hospital, Beijing, ChinaDepartment of Radiology, Peking University First Hospital, Beijing, ChinaDepartment of Radiology, Peking University First Hospital, Beijing, ChinaDepartment of Development and Research, Beijing Smart Tree Medical Technology Co. Ltd., Beijing, ChinaDepartment of Development and Research, Beijing Smart Tree Medical Technology Co. Ltd., Beijing, ChinaDepartment of Radiology, Peking University First Hospital, Beijing, ChinaPurposeTo automatically evaluate renal masses in CT images by using a cascade 3D U-Net- and ResNet-based method to accurately segment and classify focal renal lesions.Material and MethodsWe used an institutional dataset comprising 610 CT image series from 490 patients from August 2009 to August 2021 to train and evaluate the proposed method. We first determined the boundaries of the kidneys on the CT images utilizing a 3D U-Net-based method to be used as a region of interest to search for renal mass. An ensemble learning model based on 3D U-Net was then used to detect and segment the masses, followed by a ResNet algorithm for classification. Our algorithm was evaluated with an external validation dataset and kidney tumor segmentation (KiTS21) challenge dataset.ResultsThe algorithm achieved a Dice similarity coefficient (DSC) of 0.99 for bilateral kidney boundary segmentation in the test set. The average DSC for renal mass delineation using the 3D U-Net was 0.75 and 0.83. Our method detected renal masses with recalls of 84.54% and 75.90%. The classification accuracy in the test set was 86.05% for masses (<5 mm) and 91.97% for masses (≥5 mm).ConclusionWe developed a deep learning-based method for fully automated segmentation and classification of renal masses in CT images. Testing of this algorithm showed that it has the capability of accurately localizing and classifying renal masses.https://www.frontiersin.org/articles/10.3389/fonc.2023.1169922/fullrenal masscontrast-enhanced computed tomographydeep learningU-Netresidual network |
| spellingShingle | Tongtong Zhao Zhaonan Sun Ying Guo Yumeng Sun Yaofeng Zhang Xiaoying Wang Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms Frontiers in Oncology renal mass contrast-enhanced computed tomography deep learning U-Net residual network |
| title | Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms |
| title_full | Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms |
| title_fullStr | Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms |
| title_full_unstemmed | Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms |
| title_short | Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms |
| title_sort | automatic renal mass segmentation and classification on ct images based on 3d u net and resnet algorithms |
| topic | renal mass contrast-enhanced computed tomography deep learning U-Net residual network |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2023.1169922/full |
| work_keys_str_mv | AT tongtongzhao automaticrenalmasssegmentationandclassificationonctimagesbasedon3dunetandresnetalgorithms AT zhaonansun automaticrenalmasssegmentationandclassificationonctimagesbasedon3dunetandresnetalgorithms AT yingguo automaticrenalmasssegmentationandclassificationonctimagesbasedon3dunetandresnetalgorithms AT yumengsun automaticrenalmasssegmentationandclassificationonctimagesbasedon3dunetandresnetalgorithms AT yaofengzhang automaticrenalmasssegmentationandclassificationonctimagesbasedon3dunetandresnetalgorithms AT xiaoyingwang automaticrenalmasssegmentationandclassificationonctimagesbasedon3dunetandresnetalgorithms |