Skip to main content
Search

# AI in cancer detection

2 related articles

Subscribe to updates
  1. Fuzzy-rank feature selection plus H2O AutoML ensembles reach up to 95.1% accuracy and 98.1% AUC on two public cervical cancer datasets

    The work introduces an interpretable Fuzzy Rank-H2O AutoML framework in which a Fuzzy Rank Feature Selection (FRFS) algorithm picks predictors by combining statistical significance, information gain, clinical importance, and uncertainty, H2O AutoML then automatically builds ensemble models, and SHAP and LIME supply global and patient-level explanations; evaluated on two public cervical cancer datasets with stratified five-fold cross-validation where SMOTE is applied only to training folds to avoid information leakage, it reports a highest accuracy of 95.1% and AUC of 98.1%, outperforming traditional machine learning models and the baseline H2O AutoML framework.

  2. In a single-center cohort of 232 papillary thyroid carcinoma patients, an SVM model predicted lymph node metastasis with validation AUC 0.849, and removing BRAF changed AUC by only 0.007

    This retrospective study of 232 patients who underwent thyroid surgery at Tongling People's Hospital used LASSO to select six features from 32 candidates (BRAF mutation status, tumor size, capsular invasion, extrathyroidal invasion, multifocality, and TSH), compared seven machine learning algorithms, found the Support Vector Machine best in the validation cohort (AUC 0.849, 95% CI 0.756–0.934; accuracy 0.768), identified TSH and tumor size as the top SHAP contributors, and showed in an ablation analysis that removing BRAF lowered validation AUC from 0.849 to 0.843 (P = 0.671).

You’re all caught up