Table 4 Additional imaging response detection performance metrics (balanced F-scores) for each model and the two types of classifier
From: High-dimensional detection of imaging response to treatment in multiple sclerosis
Predictor set | Support vector machines | Extremely randomised trees |
|---|---|---|
Brain volume | 0.555 [0.541–0.569] | 0.601 [0.593–0.609] |
Number of lesions | 0.595 [0.581–0.609] | 0.666 [0.655–0.677] |
Total lesion volume | 0.602 [0.592–0.612] | 0.619 [0.610–0.628] |
Best low dimensional | 0.722 [0.713–0.731] | 0.717 [0.708–0.725] |
Regional atrophy | 0.849 [0.843–0.855] | 0.835 [0.829–0.842] |
Regional disconnection | 0.821 [0.814–0.829] | 0.816 [0.809–0.823] |
Best high dimensional | 0.876 [0.870–0.882] | 0.874 [0.868–0.880] |
