Machine Learning Validation
Cross-source consensus on Machine Learning Validation from 1 sources and 5 claims.
1 sources · 5 claims
Preparation
Risks & contraindications
Comparisons
Evidence quality
Highlighted claims
- Some labels were non-evaluable or unstable because class imbalance caused single-class labels or labels with only one case in a class. — Predictive radiomics for evaluation of cancer immune SignaturE in glioblastoma: the PRECISE-GBM study
- Models included support vector machines and an ensemble voting classifier combining SVM, random forest, and histogram-based gradient boosting. — Predictive radiomics for evaluation of cancer immune SignaturE in glioblastoma: the PRECISE-GBM study
- Evaluation emphasized precision, balanced accuracy, and Matthews correlation coefficient because immune labels were imbalanced and trial enrichment requires low false-positive rates. — Predictive radiomics for evaluation of cancer immune SignaturE in glioblastoma: the PRECISE-GBM study
- The study used cross-cohort external holdout strategies for analytical validation. — Predictive radiomics for evaluation of cancer immune SignaturE in glioblastoma: the PRECISE-GBM study
- Bootstrap comparisons found few statistically superior model differences after correction, although pan-cancer ensemble models tended to perform better for M0 macrophage metrics. — Predictive radiomics for evaluation of cancer immune SignaturE in glioblastoma: the PRECISE-GBM study