Table 2 The improvement of FAMO in per-case level.

From: The Feature Ambiguity Mitigate Operator model helps improve bone fracture detection on X-ray radiograph

Network

Per-case sensitivity

Per-case specificity

Youden Index (%)

AUC (%)

ResNeXt101+FPN

349/466

(74.9%)

443/483

(91.7%)

66.6

85.7

CI: (70.6%, 78.7%)

CI: (88.8%, 93.9%)

CI: (59.4, 73.8)

CI: (84.8, 86.5)

FAMO+ResNext101+FPN

361/466

(77.5%)

451/484

(93.4%)

70.9

86.5

CI: (73.3%, 81.1%)

CI: (90.7%, 95.4%)

CI: (64.0, 77.8)

CI:(85.6, 87.4)

  1. ResNeXt101, a network structure; FPN, feature pyramid network; FAMO, Feature Ambiguity Mitigate Operator (FAMO) method; AUC, area under the curve; AP, average precision; CI, confidence interval. All CIs were calculated at the two-tailed significance of 0.05.
Back to article page