Artificial Intelligence-Based Diagnosis of Gastric Mesenchymal Tumors Using Digital Endosonography Image Analysis
Source check: Source checked. Not independent scientific confirmation.
Finding: Clinical/endoscopic factors improved random-forest and KNN discrimination of GIST versus non-GIST, with AUCs 0.896 and 0.825. Retrospective, single-center, 387 analyzed resection cases with histopathologic labels and 10-fold cross-validation; selection bias toward surgically resected tumors and lack of independent external validation are explicit limitations.
Technology: Logistic regression, decision tree, random forest, KNN, and SVM classifiers use digital EUS image features; combined models add age and tumor location.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Logistic regression, decision tree, random forest, KNN, and SVM classifiers use digital EUS image features; combined models add age and tumor location.
- Test and result
- Clinical/endoscopic factors improved random-forest and KNN discrimination of GIST versus non-GIST, with AUCs 0.896 and 0.825.
- Limitation / still unknown
- Retrospective, single-center, 387 analyzed resection cases with histopathologic labels and 10-fold cross-validation; selection bias toward surgically resected tumors and lack of independent external validation are explicit limitations.
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/PMC11242784/
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 source full text Methods and Results; version/date checked against bibliographic record..
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 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.