PET radiomics classifies pathological prostate cancer risk
Source check: Source checked. Not independent scientific confirmation.
Finding: Random forests analyzed preoperative PSMA PET images from 76 patients with prostate cancer against pathology-confirmed disease features. Repeated internal cross-validation reported AUC 0.86 for lymph-node involvement and 0.81 for high Gleason grade. Radiomics had higher mean scores than standard PET measures, but the difference was not statistically significant. This does not establish improved care or external clinical performance.
Technology: Random forest; nested feature selection
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Random forests learn relationships between PET intensity, shape and texture measurements and pathological risk categories.
- Test and result
- Pathology-confirmed endpoints in a 76-person cohort, with repeated five-fold cross-validation, training-fold preprocessing and nested model tuning.
- Limitation / still unknown
- Multicentre frozen-model validation and clinical benefit remain unestablished; the small cohort and selection among analysis configurations limit generalizability.
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/PMC7835295/
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 relevant model Methods, Results and limitations; exact primary identity and current primary PubMed correction/retraction relationships checked..
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.