Table 3 The AUC results of the testing set obtained through 100 iterations of data shuffling.

From: A multimodal machine learning model for predicting dementia conversion in Alzheimer’s disease

Modality combination

Machine learning models

DT

RF

SVM

LR

GBM

XGB

Mean ± SD

Mean ± SD

Mean ± SD

Mean ± SD

Mean ± SD

Mean ± SD

demo

0.647 ± 0.093

0.672 ± 0.084

0.658 ± 0.091

0.722 ± 0.073

0.655 ± 0.093

0.679 ± 0.086

A

0.698 ± 0.094

0.788 ± 0.072

0.762 ± 0.080

0.756 ± 0.078

0.757 ± 0.076

0.792 ± 0.066

N

0.658 ± 0.091

0.763 ± 0.080

0.737 ± 0.066

0.648 ± 0.071

0.782 ± 0.080

0.766 ± 0.084

V

0.471 ± 0.096

0.495 ± 0.084

0.489 ± 0.088

0.558 ± 0.094

0.479 ± 0.079

0.467 ± 0.083

demo + A

0.722 ± 0.094

0.813 ± 0.073

0.788 ± 0.071

0.796 ± 0.071

0.803 ± 0.070

0.826 ± 0.063

demo + N

0.661 ± 0.093

0.782 ± 0.081

0.759 ± 0.063

0.644 ± 0.072

0.799 ± 0.071

0.780 ± 0.077

demo + V

0.610 ± 0.107

0.619 ± 0.086

0.646 ± 0.099

0.705 ± 0.091

0.599 ± 0.083

0.619 ± 0.084

demo + AN

0.722 ± 0.091

0.844 ± 0.063

0.826 ± 0.056

0.796 ± 0.064

0.881 ± 0.053

0.865 ± 0.057

demo + NV

0.667 ± 0.101

0.781 ± 0.080

0.758 ± 0.067

0.647 ± 0.073

0.798 ± 0.075

0.779 ± 0.083

demo + AV

0.716 ± 0.091

0.807 ± 0.072

0.778 ± 0.070

0.809 ± 0.067

0.803 ± 0.068

0.821 ± 0.061

demo + ANV

0.728 ± 0.098

0.840 ± 0.063

0.816 ± 0.055

0.799 ± 0.065

0.879 ± 0.053

0.863 ± 0.064

  1. AUC; Area Under ROC Curve, SD; Standard Deviation, CI; Confidence Interval, DT; Decision Trees, RF; Random Forests, SVM; Support Vector Machines, LR; Linear Regression Classifiers, GBM; Gradient Boosting Models, XGB; Extreme Gradient Boosting, demo; demographic characteristics, A; amyloid PET image features, N; T1-weigted image features, V; T2-FLAIR image features.
Back to article page