Table 4 Results on the ADNI2 & ADNI3 datasets. The numbers indicate out-of-fold means and standard deviation. The performance of the model is evaluated using the following metrics: balanced accuracy (BACC), sensitivity (SEN), specificity (SPE), and area under the curve (AUC).

CN vs. AD

MRI

\(0.851 \pm 0.052\)

\(0.869 \pm 0.076\)

\(0.832 \pm 0.105\)

\(0.866 \pm 0.048\)

A\(\beta\)

\(0.856 \pm 0.068\)

\(0.890 \pm 0.133\)

\(0.821 \pm 0.099\)

\(0.866 \pm 0.061\)

Tau

\(0.907 \pm 0.051\)

\(0.853 \pm 0.092\)

\(0.961 \pm 0.024\)

\(0.929 \pm 0.037\)

MRI,A\(\beta\)

\(0.856 \pm 0.076\)

\(0.795 \pm 0.137\)

\(0.918 \pm 0.031\)

\(0.871 \pm 0.066\)

MRI,Tau

\(0.904 \pm 0.054\)

\(0.871 \pm 0.092\)

\(0.938 \pm 0.060\)

\(0.893 \pm 0.082\)

A\(\beta\),Tau

\(0.877 \pm 0.069\)

\(0.796 \pm 0.145\)

\(0.958 \pm 0.046\)

\(0.894 \pm 0.065\)

MRI,A\(\beta\),Tau

\(\mathbf {0.936} \pm \mathbf {0.039}\)

\(\mathbf {0.907} \pm \mathbf {0.058}\)

\(\mathbf {0.965} \pm \mathbf {0.031}\)

\(\mathbf {0.949} \pm \mathbf {0.033}\)

CN vs. MCI

MRI

\(0.660 \pm 0.029\)

\(0.622 \pm 0.136\)

\(0.697 \pm 0.102\)

\(0.661 \pm 0.037\)

A\(\beta\)

\(0.708 \pm 0.053\)

\(0.598 \pm 0.147\)

\(0.817 \pm 0.097\)

\(0.702 \pm 0.109\)

Tau

\(0.712 \pm 0.029\)

\(0.598 \pm 0.055\)

\(0.825 \pm 0.052\)

\(0.758 \pm 0.060\)

MRI,A\(\beta\)

\(0.717 \pm 0.063\)

\(0.637 \pm 0.168\)

\(0.797 \pm 0.080\)

\(0.731 \pm 0.075\)

MRI,Tau

\(0.741 \pm 0.032\)

\(0.644 \pm 0.127\)

\(0.837 \pm 0.107\)

\(0.768 \pm 0.042\)

A\(\beta\),Tau

\(0.727 \pm 0.044\)

\(0.575 \pm 0.063\)

\(\mathbf {0.880} \pm \mathbf {0.045}\)

\(0.746 \pm 0.075\)

MRI,A\(\beta\),Tau

\(\mathbf {0.753} \pm \mathbf {0.036}\)

\(\mathbf {0.716} \pm \mathbf {0.047}\)

\(0.790 \pm 0.046\)

\(\mathbf {0.793} \pm \mathbf {0.037}\)