Table 1.

Model input variables and abbreviations for machine-learning classifiers and feature selection methods.

A) Clinical Variables	B) Treatment Variable
Age	Thrombectomy reperfusion success: modified treatment in cerebral ischemia (mTICI) score
Sex
Admission NIH Stroke Score	Intravenous thrombolytic therapy
C) Machine-Learning Classifiers
Random forest	RF
XGBoost	XGB
Logistic regression with elastic net regularization	ElNet
Native Bayes classifier	NBayes
Support vector machine with radial kernel	SVM_rad
Support vector machine with sigmoid kernel	SVM_sig
D) Feature Selection Methods
Minimum redundancy maximum relevance filter	MRMR
Pearson correlation-based redundancy reduction combined with a mutual information maximization filter	pMIM
Logistic regression with RIDGE regularization adapted for feature selection	RIDGE
Hierarchical clustering	HClust
Principal component analysis-based feature selection	PCA
No feature selection implemented	noFS

Three main prognostic clinical variables at the time of admission (A) were included in the Combined and Clinical + Treament models. The treatment variables of post-thrombectomy reperfusion success (mTICI ascore) and intravenous thrombolytic treatment (B) were used in the Radiomics + Treatment, Clinical + Treatment, and Combined models. Six machine-learning classifiers (C) and 6 feature selection methods (D) were used in 36 combinations for the Radiomics, Radiomics + Treatment, Combined models, while feature selection was omitted in Clinical + Treatment models. Machine-learning and feature selection abbreviations were previsouly described in detail (Haider et al., 2020b) and are summarized in the supplementary methods.