Skip to main content

Table 3 Machine learning classification algorithms evaluated in this work, together with their associated parameters and notation

From: Comparing radiomic classifiers and classifier ensembles for detection of peripheral zone prostate tumors on T2-weighted MRI: a multi-site study

QDA [23]

h QDA

-

MATLAB

 

hBag, QDA, hBoost, QDA

T=50

MATLAB

SVM [24, 49]

h SVM

Ω,λ

LIBSVM

 

hBag, SVM, hBoost, SVM

Ω,λ,T=50

LIBSVM [49], MATLAB

Naïve Bayes [50]

h Bay

-

MATLAB

 

hBag, Bay, hBoost, Bay

T=50

MATLAB

Decision Trees [26]

h DT

-

C4.5

 

hBag, DT, hBoost, DT

T=50

MATLAB TreeBagger, PBTs [51]

  1. SVM parameters include Ω (trade-off between training error and model complexity) and λ (normalization factor for inputs), which are determined via a grid search strategy. For ensemble approaches, T refers to the number of component classifiers