Skip to main content

Table 4 Bootstrap test of ROC between 3D-Mask-RCNN and 3D-Unet

From: Automated meniscus segmentation and tear detection of knee MRI with a 3D mask-RCNN

ROC

FOM

CI

p-value

3D-Mask-RCNN

0.4237

(0.1075,0.8406)

0.0009

Radiological evaluation

  1. FOM is the difference in the area under FROC curve between the two methods at a threshold of 4 FPs/view; CI confidence interval
Back to article page

European Journal of Medical Research

ISSN: 2047-783X

Contact us