Skip to main content
. 2022 Feb 23;13:819055. doi: 10.3389/fneur.2022.819055

Table 3.

The statistics for evaluating classification performance.

Modality ACC SEN SPE AUC P
Condition:CSVD-c group vs. control group
GMV 84.62% 76.92% 89.74% 0.857 <0.05
WMV 76.92% 69.23% 82.05% 0.817 <0.05
GMV+WMV 86.15% 80.77% 89.74% 0.926 <0.05
Condition:CSVD-n group vs. control group
GMV 74.39% 67.44% 82.05% 0.815 <0.05
WMV 74.39% 65.12% 84.62% 0.804 <0.05
GMV+WMV 81.71% 69.77% 94.87% 0.891 <0.05
Condition:CSVD-c group vs. CSVD-n group
GMV 68.12% 57.69% 74.42% 0.770 <0.05
WMV 79.71% 73.08% 83.72% 0.815 <0.05
GMV+WMV 81.16% 69.23% 88.37% 0.881 <0.05

ACC/SEN/SPE, accuracy/sensitivity/specificity; AUC, area under the ROC curve. P: p-values generated by non-parametric permutation testing with 5,000 randomizations.