Digital kidney-biopsy damage features are associated with later graft loss
Source check: Source checked. Not independent scientific confirmation.
Finding: Deep-learning image analysis quantified tissue damage in 789 kidney-transplant biopsy slides from two cohorts. A damage score from 12-month biopsies was associated with subsequent graft loss in both cohorts. Baseline-score survival association was not confirmed in the smaller Australian cohort, and clinical utility remains untested.
Technology: Mask R-CNN and U-Net histopathology analysis
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Neural tissue/immune-cell segmentation supplied quantitative damage features, which ordinary survival analysis related to subsequent graft outcomes.
- Test and result
- Networks were developed on 60 selected biopsy slides and tested on 33 others. The pipeline then analyzed 478 baseline and 311 twelve-month slides from 404 GoCAR and 212 AUSCAD participants. In 200 GoCAR and 111 AUSCAD twelve-month biopsies, composite tissue/inflammation scores were associated with later death-censored graft loss; thresholds were derived in GoCAR.
- Limitation / still unknown
- This is retrospective image/outcome analysis of previously collected cohorts, not prospective evaluation of a frozen clinical tool. GoCAR supplies both selected model-development slides and outcome analyses, with exact patient overlap not separated here. Baseline ITAS survival association was not confirmed in AUSCAD. Observational associations, selected thresholds, limited graft-loss events and cohort differences prevent causal, treatment-benefit or universal surrogate-endpoint claims.
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/PMC10285669/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 21 Nov 2026. Review scope: Primary full-text bibliographic dates, computational classification Methods, held-out Results and explicit cohort/reference limitations..
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 1/5 · 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.