Supplementary Materials for "Explainable AI for Lower-Grade Glioma and Glioblastoma Classification: Model Comparison and Feature Selection"
Year: 2026
Type: supplementary-materials
Abstract: This record contains six supplementary CSV tables accompanying the manuscript “Explainable AI for Lower-Grade Glioma and Glioblastoma Classification: Model Comparison and Feature Selection.” The files provide additional details on the model search and evaluation: Supplementary Table S1 reports hyperparameter search spaces and implementation details. Supplementary Table S2 presents corrected paired differences and equivalence screening for the reduced-feature configurations. Supplementary Table S3 provides fold-wise differences for the selected seven-feature Random Forest model and its full-feature reference. Supplementary Table S4 compares full-feature Random Forest ROC-AUC results across two experiments. Supplementary Table S5 reports the stability of SHAP feature rankings across outer training folds. Supplementary Table S6 summarizes patient profiles by out-of-fold prediction outcome. The analyses use the Glioma Grading Clinical and Mutation Features dataset from the UCI Machine Learning Repository, derived from the TCGA-LGG and TCGA-GBM cohorts. These files supplement the results reported in the manuscript.
Author: Tymoshchuk, Dmytro
Institution: Ternopil Ivan Pului National Technical University
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7225323441