Table 2 Comparison of the performance between Transformer-based model and conventional machine learning models

From: Deep adaptive learning predicts and diagnoses CSVD-related cognitive decline using radiomics from T2-FLAIR: a multi-centre study

 

AUC

Accuracy

Sensitivity

Specificity

Precision

Recall

Transformer

0.841 ± 0.016

0.798 ± 0.021

0.793 ± 0.108

0.800 ± 0.065

0.716 ± 0.055

0.793 ± 0.108

Random Forest

0.820 ± 0.025

0.747 ± 0.008

0.697 ± 0.052

0.777 ± 0.027

0.665 ± 0.046

0.697 ± 0.052

SVM

0.770 ± 0.031

0.712 ± 0.015

0.633 ± 0.081

0.759 ± 0.047

0.627 ± 0.044

0.633 ± 0.081

XGBoost

0.813 ± 0.033

0.724 ± 0.032

0.672 ± 0.070

0.754 ± 0.029

0.633 ± 0.063

0.672 ± 0.070

  1. AUC area under the curve, SVM Support Vector Machine.
Back to article page