Fully automated segmentation and classification of renal tumors on CT scans via machine learning
Abstract Background To develop and test the performance of a fully automated system for classifying renal tumor subtypes via deep machine learning for automated segmentation and classification. Materials and methods The model was developed using computed tomography (CT) images of pathologically prov...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BMC Cancer |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12885-025-13582-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571616064700416 |
|---|---|
| author | Jang Hee Han Byung Woo Kim Taek Min Kim Ji Yeon Ko Seung Jae Choi Minho Kang Sang Youn Kim Jeong Yeon Cho Ja Hyeon Ku Cheol Kwak Young-Gon Kim Chang Wook Jeong |
| author_facet | Jang Hee Han Byung Woo Kim Taek Min Kim Ji Yeon Ko Seung Jae Choi Minho Kang Sang Youn Kim Jeong Yeon Cho Ja Hyeon Ku Cheol Kwak Young-Gon Kim Chang Wook Jeong |
| author_sort | Jang Hee Han |
| collection | DOAJ |
| description | Abstract Background To develop and test the performance of a fully automated system for classifying renal tumor subtypes via deep machine learning for automated segmentation and classification. Materials and methods The model was developed using computed tomography (CT) images of pathologically proven renal tumors collected from a prospective cohort at a medical center between March 2016 and December 2020. A total of 561 renal tumors were included: 233 clear cell renal cell carcinomas (RCCs), 82 papillary RCCs, 74 chromophobe RCCs, and 172 angiomyolipomas. Renal tumor masks manually drawn on contrast-enhanced CT images were used to develop a 3D U-Net-based deep learning model for fully automated tumor segmentation. After segmentation, the entire classification pipeline, including feature extraction and subtype classification, was conducted without any manual intervention. Both conventional radiological features (Hounsfield units, HUs) and radiomic features extracted from areas predicted by the deep learning models were used to develop an algorithm for classifying renal tumor subtypes via a random forest classifier. The performance of the segmentation model was evaluated using the Dice similarity coefficient, while the classification model was assessed based on accuracy, sensitivity, and specificity. Results For tumors larger than 4 cm, the Dice similarity coefficient (DSC) for automated segmentation was 0.83, while for tumors smaller than 4 cm, the DSC was 0.65. The classification accuracy (ACC) for distinguishing RCC subtypes was 0.77 for tumors larger than 4 cm and 0.68 for tumors smaller than 4 cm. Additionally, the accuracy for benign versus malignant classification was 0.85. Conclusions Our automatic segmentation and classifier model showed promising results for renal tumor segmentation and classification. |
| format | Article |
| id | doaj-art-cf95ad8a946d41009640b923dcc09928 |
| institution | Kabale University |
| issn | 1471-2407 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Cancer |
| spelling | doaj-art-cf95ad8a946d41009640b923dcc099282025-02-02T12:28:49ZengBMCBMC Cancer1471-24072025-01-012511910.1186/s12885-025-13582-6Fully automated segmentation and classification of renal tumors on CT scans via machine learningJang Hee Han0Byung Woo Kim1Taek Min Kim2Ji Yeon Ko3Seung Jae Choi4Minho Kang5Sang Youn Kim6Jeong Yeon Cho7Ja Hyeon Ku8Cheol Kwak9Young-Gon Kim10Chang Wook Jeong11Department of Urology, Seoul National University HospitalDepartment of Transdisciplinary Medicine, Seoul National University HospitalDepartment of Radiology, Seoul National University HospitalDepartment of Urology, Seoul National University HospitalDepartment of Transdisciplinary Medicine, Seoul National University HospitalDepartment of Urology, Seoul National University HospitalDepartment of Radiology, Seoul National University HospitalDepartment of Radiology, Seoul National University HospitalDepartment of Urology, Seoul National University HospitalDepartment of Urology, Seoul National University HospitalDepartment of Transdisciplinary Medicine, Seoul National University HospitalDepartment of Urology, Seoul National University HospitalAbstract Background To develop and test the performance of a fully automated system for classifying renal tumor subtypes via deep machine learning for automated segmentation and classification. Materials and methods The model was developed using computed tomography (CT) images of pathologically proven renal tumors collected from a prospective cohort at a medical center between March 2016 and December 2020. A total of 561 renal tumors were included: 233 clear cell renal cell carcinomas (RCCs), 82 papillary RCCs, 74 chromophobe RCCs, and 172 angiomyolipomas. Renal tumor masks manually drawn on contrast-enhanced CT images were used to develop a 3D U-Net-based deep learning model for fully automated tumor segmentation. After segmentation, the entire classification pipeline, including feature extraction and subtype classification, was conducted without any manual intervention. Both conventional radiological features (Hounsfield units, HUs) and radiomic features extracted from areas predicted by the deep learning models were used to develop an algorithm for classifying renal tumor subtypes via a random forest classifier. The performance of the segmentation model was evaluated using the Dice similarity coefficient, while the classification model was assessed based on accuracy, sensitivity, and specificity. Results For tumors larger than 4 cm, the Dice similarity coefficient (DSC) for automated segmentation was 0.83, while for tumors smaller than 4 cm, the DSC was 0.65. The classification accuracy (ACC) for distinguishing RCC subtypes was 0.77 for tumors larger than 4 cm and 0.68 for tumors smaller than 4 cm. Additionally, the accuracy for benign versus malignant classification was 0.85. Conclusions Our automatic segmentation and classifier model showed promising results for renal tumor segmentation and classification.https://doi.org/10.1186/s12885-025-13582-6CTRenal cell carcinomaRadiomicsMachine learningDeep learningSegmentation |
| spellingShingle | Jang Hee Han Byung Woo Kim Taek Min Kim Ji Yeon Ko Seung Jae Choi Minho Kang Sang Youn Kim Jeong Yeon Cho Ja Hyeon Ku Cheol Kwak Young-Gon Kim Chang Wook Jeong Fully automated segmentation and classification of renal tumors on CT scans via machine learning BMC Cancer CT Renal cell carcinoma Radiomics Machine learning Deep learning Segmentation |
| title | Fully automated segmentation and classification of renal tumors on CT scans via machine learning |
| title_full | Fully automated segmentation and classification of renal tumors on CT scans via machine learning |
| title_fullStr | Fully automated segmentation and classification of renal tumors on CT scans via machine learning |
| title_full_unstemmed | Fully automated segmentation and classification of renal tumors on CT scans via machine learning |
| title_short | Fully automated segmentation and classification of renal tumors on CT scans via machine learning |
| title_sort | fully automated segmentation and classification of renal tumors on ct scans via machine learning |
| topic | CT Renal cell carcinoma Radiomics Machine learning Deep learning Segmentation |
| url | https://doi.org/10.1186/s12885-025-13582-6 |
| work_keys_str_mv | AT jangheehan fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT byungwookim fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT taekminkim fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT jiyeonko fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT seungjaechoi fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT minhokang fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT sangyounkim fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT jeongyeoncho fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT jahyeonku fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT cheolkwak fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT younggonkim fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning AT changwookjeong fullyautomatedsegmentationandclassificationofrenaltumorsonctscansviamachinelearning |