🤖 Generated with AI assistance · Does not replace individualized medical advice.
Source: PubMed PMID 42670720 · BJU international (2026)
Score: 5.0/10 · Generated: 2026-09-01T07:26:44.498293 · Engine: qwen36-dsv4 local
This content was generated with artificial intelligence assistance and reviewed by a medical oncologist. It does not constitute medical advice.
Predicting the composition of kidney stones using artificial intelligence could transform clinical management. A recent study shows that machine learning models, especially when incorporating morphological features, achieve exceptional accuracy.
Keyfacts
| Concept | Detail |
|---|---|
| Design | Retrospective cohort, 2019–2024 PMID: 42670720 |
| Patients | 442 (median age 51 years, 67% men) PMID: 42670720 |
| Models | 8 supervised algorithms (logistic regression, SVM, Random Forest, ExtraTrees, AdaBoost, XGBoost, CatBoost) PMID: 42670720 |
| Performance | macro-AUC up to 0.983 with CatBoost when adding morphology [PMID: 42670720 |