Table 2.

(A,B) The ADHD-TDC classification accuracy of models based on LBP-TOP features with different registration methods, parcellation, and radius of LBP-TOP, using 1NN and linear-SVM classifiers alternatively. (C,D) The ADHD-TDC classification accuracy of models based on simple rs-fMRI features, using 1NN and linear-SVM classifiers alternatively.

The highest accuracy for each parcellation is denoted by the bold number. Abbreviations are as follows: R1, R2, and R3, LBP-TOP radius in mm; DOF9, and DOF12, linear registration with 9, and 12 degree of freedom, respectively; ART, non-linear registration performed by Automated Registration Tool; AAL, automated anatomical labeling template; and CC200, spatially constrained parcellation based on rs-fMRI. Abbreviations are as follows: BF, rs-fMRI data filtered by a bandpass filter (0.009 Hz ~ 0.08 Hz); and non-BF, rs-fMRI data not filtered by a bandpass filter.