CT-based clinical deep learning radiomics nomogram for preoperative differentiation of pyelocaliceal upper tract urothelial carcinoma and xanthogranulomatous pyelonephritis
Source check: Source checked. Not independent scientific confirmation.
Finding: A CT radiomics and deep-learning nomogram was evaluated to distinguish upper-tract urothelial carcinoma from xanthogranulomatous pyelonephritis in 161 patients with pathology-confirmed diagnoses. On an internal random split, the combined model had validation AUC 0.930 versus 0.695 for a radiologist. External validation and clinical utility remain unestablished.
Technology: CT radiomics and deep-learning nomogram
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Radiomics and deep-learning features from preoperative CT were combined to distinguish upper-tract urothelial carcinoma from xanthogranulomatous pyelonephritis.
- Test and result
- In 161 patients with pathology-confirmed UTUC or XGP, a random 112/49 training/validation split was used. The clinical deep-learning radiomics model had validation AUC 0.930 and exceeded a radiologist’s AUC of 0.695 in this cohort.
- Limitation / still unknown
- The evidence is internal validation in a limited cohort; external validation and clinical utility are not established.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.3389/fonc.2026.1880690
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.