CT-based clinical deep learning radiomics nomogram for preoperative differentiation of pyelocaliceal upper tract urothelial carcinoma and xanthogranulomatous pyelonephritis
Frontiers in Oncology · 02.11.2026 · 16
DOI: 10.3389/fonc.2026.1880690 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Oncology, online erschienen am 02.11.2026. Verfasst von Yi Feng und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Urology, the First Affiliated Hospital of Guangxi Medical University, Nanning, China. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective
This study sought to create a CT-based clinical deep learning radiomics (CDLR) model to noninvasively differentiate between pyelocaliceal upper tract urothelial carcinoma (UTUC) and xanthogranulomatous pyelonephritis (XGP) preoperatively, aiding in personalized treatment decisions.
Methods
A retrospective analysis included 161 patients with pathologically confirmed pyelocaliceal UTUC or XGP following surgery. Patients were randomly divided into a training cohort (n=112) and a validation cohort (n=49). Radiomics (Rad) features and deep learning (DL) features were extracted from the corticomedullary phase CT images. Combined with clinical features, and after feature selection, a logistic regression (LR) classifier was employed to construct the clinical model, the deep learning radiomics (DLR) model, and the clinical deep learning radiomics (CDLR) model. Model performance was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA), with interpretability assessed by the shapley additive explanations (SHAP) method. In addition, the diagnostic performance of the models was compared with that of a radiologist with 5 years of experience in abdominal CT diagnosis.
Results
The CDLR model achieved AUCs of 0.955 in the training cohort and 0.930 in the validation cohort. In the validation cohort, the CDLR model significantly outperformed the radiologist’s assessment (AUC: 0.930 vs. 0.695, P = 0.0006) and yielded numerically higher AUCs than the clinical model (0.823) and the DLR model (0.882). DCA indicated that the CDLR model provided the highest net clinical benefit. SHAP analysis demonstrated good model interpretability.
Conclusion
The CDLR model integrating clinical, Rad, and DL features serves as an effective noninvasive tool for preoperative differentiation of pyelocaliceal UTUC and XGP, offering valuable reference for clinical decision-making.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Urology, the First Affiliated Hospital of Guangxi Medical University, Nanning, China.
- Department of Radiology, the First Affiliated Hospital of Guangxi Medical University, Nanning, China.
- Medical College of Oncology, Guangxi Medical University, Nanning, China.
- School of Basic Medical Sciences, Guangxi Medical University, Nanning, China.
- Department of Urology, The First People's Hospital of Yulin, Yulin, China.
Zitieren
APA
Feng, Y., Cheng, Y., Zeng, J., Lin, J., Qin, H., Yan, Y., Huang, Z., Wu, Y., Liu, X., Fang, F., Mo, L., Xu, W., & Li, C. (2026). CT-based clinical deep learning radiomics nomogram for preoperative differentiation of pyelocaliceal upper tract urothelial carcinoma and xanthogranulomatous pyelonephritis. Frontiers in Oncology, 16. https://doi.org/10.3389/fonc.2026.1880690
Vancouver
Feng Y, Cheng Y, Zeng J, Lin J, Qin H, Yan Y, et al. CT-based clinical deep learning radiomics nomogram for preoperative differentiation of pyelocaliceal upper tract urothelial carcinoma and xanthogranulomatous pyelonephritis. Frontiers in Oncology. 2026;16. doi:10.3389/fonc.2026.1880690
BibTeX
@article{Feng2026,
author = {Feng, Yi and Cheng, Yiheng and Zeng, Jian and Lin, Junyu and Qin, Hao and Yan, Yizheng and Huang, Zile and Wu, Yongxian and Liu, Xiao and Fang, Fuxiang and Mo, Linjian and Xu, Wei and Li, Chengyang},
title = {CT-based clinical deep learning radiomics nomogram for preoperative differentiation of pyelocaliceal upper tract urothelial carcinoma and xanthogranulomatous pyelonephritis},
journal = {Frontiers in Oncology},
year = {2026},
volume = {16},
doi = {10.3389/fonc.2026.1880690}
}
EUVIMED