AutoParis-X quantifies longitudinal urine-cytology signals of bladder-cancer recurrence
Source check: Source checked. Not independent scientific confirmation.
Finding: AutoParis-X extracted cell and cluster features from 1,259 urine specimens from 135 bladder-cancer patients to explore longitudinal recurrence markers. Quantitative predictors were associated with recurrence, with reported concordance increasing from 0.672 at the initial diagnosis window to 0.77 when collecting information through 180 days. This exploratory retrospective study accounted for delayed entry but lacked external validation; treatment recording, sporadic follow-up and competing mortality limited inference. Clinical decision benefits remain untested.
Technology: AutoParis-X; UroNet; AtyNet
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- UroNet CNN filters urothelial cells, AtyNet CNN scores cellular atypia and a learned classifier integrates atypia burden; features enter time-varying Cox and trajectory analyses. Source reports 135 patients before exclusions; do not equate this with every analysis denominator.
- Test and result
- Exploratory retrospective cytology/recurrence associations and comparison with manual cytology and histology.
- Limitation / still unknown
- External replication, reliable treatment adjustment and benefits of decision support remain unestablished.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC10527805/
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: 21 Nov 2026. Review scope: Primary full article Methods, Results and limitations plus Europe PMC bibliographic metadata; PMC full article..
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.