Weakly human‐supervised deep learning for real‐time detection of high‐grade aggressive clear cell renal cell carcinoma on contrast‐enhanced CT
Medical Physics · 2026
DOI: 10.1002/mp.70348 ↗At a glance
Journal article in Medical Physics, 2026. Written by Rui Zhi and 8 co-authors. Institution: Department of Radiology, The First Affiliated Hospital with Nanjing Medical University, Nanjing, Jiangsu Province, P.R. China. Major MEDLINE topics: Carcinoma, Renal Cell, Tomography, X-Ray Computed, Contrast Media, Supervised Machine Learning. Cited 1 times so far (Crossref); 48 references. Funded by: National Natural Science Foundation of China, Zhejiang Medical and Health Science and Technology Program.
Abstract (original language)
Background
The performance of contrast‐enhanced computed tomography (CECT) in staging clear cell renal cell carcinomas (ccRCCs) and assessing tumor aggressiveness remains limited by heterogeneous and poor sensitivity.
Purpose
This study aims to design and validate a human‐AI interactive network, Kidney Tumor Staging Network (KtSNet), which leverages weakly supervised learning for real‐time, efficient detection of high‐grade aggressive ccRCC (HGRCC) using CECT. Materials and methods A total of 1,092 patients with ccRCC were enrolled across five cohort datasets (training/internal testing/external testing, n = 611/153/328). To achieve precise pre‐surgical detection of HGRCC on CT imaging, we pretrained a self‐supervised foundation model (SSFM) using a large cross‐modal dataset ( n = 40 000) for image restoration‐based transfer learning. To develop human‐AI interactive capabilities, we trained KtSNet by integrating SSFM with weakly supervised learning, enabling real‐time determination of HGRCC on CT imaging through human‐AI interaction.
Results
In the internal test cohort comprising 153 patients, KtSNet demonstrated significantly higher Area Under the Curve (AUC) values for both ROC and PR curves (ROC‐AUC = 0.76; PR‐AUC = 0.29, F 1 max : 0.441) compared to B2Net (ROC‐AUC = 0.68, p = 0.040; PR‐AUC = 0.22, F 1 max : 0.366), RML‐XGB (ROC‐AUC = 0.53, p < 0.001; PR‐AUC = 0.14, F 1 max : 0.264), and Likert scoring (ROC‐AUC values of 0.57, 0.58, and 0.70 for the three readers) in staging HGRCC. In the external validation cohort, KtSNet maintained superior AUCs on both ROC and PR curves (ROC‐AUC = 0.85; PR‐AUC = 0.42, F 1 max : 0.529) compared to B2Net (ROC‐AUC = 0.74, p = 0.002; PR‐AUC = 0.21, F 1 max : 0.328) and exhibited significantly higher AUCs than RML‐XGB (ROC‐AUC = 0.63, p = 0.002; PR‐AUC = 0.23, F 1 max : 0.359).
Conclusions
The weakly human‐supervised KtSNet may serve as a promising opportunity for real‐time determination of HGRCC using CT imaging.
Subjects
MeSH terms (MEDLINE)
Author keywords
Substances
Authors and institutions
- Department of Radiology, The First Affiliated Hospital with Nanjing Medical University, Nanjing, Jiangsu Province, P.R. China.
- Department of Urology, Hangzhou Hospital of Traditional Chinese Medicine, Hangzhou, Zhejiang Province, P.R. China.
- Department of Urology, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, P. R. China.
- Department of Urology, The Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, P.R. China.
- Jiangsu Province (Suqian) Hospital, Suqian First Hospital, Suqian, Jiangsu Province, China.
- Department of Radiology, Peking University First Hospital, Beijing, P.R. China.
Funding
- National Natural Science Foundation of China
- Zhejiang Medical and Health Science and Technology Program
Cite
APA
Zhi, R., Li, Q., Shan, S., Gao, F., Huang, J., Zhang, N., Cai, W., Zhang, Y. D., & Zhang, X. (2026). Weakly human‐supervised deep learning for real‐time detection of high‐grade aggressive clear cell renal cell carcinoma on contrast‐enhanced CT. Medical Physics. https://doi.org/10.1002/mp.70348
Vancouver
Zhi R, Li Q, Shan S, Gao F, Huang J, Zhang N, et al. Weakly human‐supervised deep learning for real‐time detection of high‐grade aggressive clear cell renal cell carcinoma on contrast‐enhanced CT. Medical Physics. 2026. doi:10.1002/mp.70348
BibTeX
@article{Zhi2026,
author = {Zhi, Rui and Li, Qiao and Shan, Shuai and Gao, Feng and Huang, Jiwei and Zhang, Nan and Cai, Wen and Zhang, Yu-Dong and Zhang, Xiaodong},
title = {Weakly human‐supervised deep learning for real‐time detection of high‐grade aggressive clear cell renal cell carcinoma on contrast‐enhanced CT},
journal = {Medical Physics},
year = {2026},
doi = {10.1002/mp.70348}
}
EUVIMED