Table 12.

Multi-organ lesion detection (MOLD) results of DA-SHT Dense 3DCE R-FCN, 9 slices model on the extracted multi-organ DeepLesion dataset using different values of $\beta$ in Equation (1). The default setting of using $\alpha$ = 10 is the same as that in [1,25]. AS (%) at various FPs per image and mAP (%) were used as the evaluation metrics. $std$ denotes the standard deviation value.

$\mathit{\alpha }$ and $\mathit{\beta }$ Value	AS@0.5 ± std	AS@1 ± std	AS@2 ± std	AS@4 ± std	AS@8 ± std	AS@16 ± std	mAP ± std
$\alpha$ = 10, $\beta$ = 0.1	35.50 ± 0.22	47.13 ± 0.67	58.00 ± 0.04	66.69 ± 0.06	74.06 ± 0.52	77.83 ± 0.29	31.96 ± 0.15
$\alpha$ = 10, $\beta$ = 1	44.37 ± 0.12	58.05 ± 1.13	67.29 ± 0.12	73.88 ± 0.17	78.97 ± 0.14	82.11 ± 0.12	40.39 ± 0.41
$\alpha$ = 10, $\beta$ = 10	36.20 ± 0.19	48.48 ± 0.46	60.04 ± 0.73	67.04 ± 0.79	72.25 ± 0.81	77.51 ± 0.81	32.24 ± 0.37
$\alpha$ = 10, $\beta$ = 100	37.60 ± 0.17	49.43 ± 0.52	60.44 ± 0.33	68.10 ± 0.48	75.83 ± 0.33	79.59 ± 0.20	34.42 ± 0.24