Skip to main content

Table 2 Mean values of the metrics for the dataset B, using the validation set and cropped images resized to 160 × 160 pixels

From: Evaluating the performance of convolutional neural networks with direct acyclic graph architectures in automatic segmentation of breast lesion in US images

CNN

Global Accuracy

Accuracy

IoU

Weighted IoU

BF Score

Dice Coefficient

CNN1

 Mean

0.917

0.904

0.836

0.850

0.510

0.920

 Standard Deviation

0.048

0.068

0.090

0.079

0.083

0.032

CNN2

 Mean

0.911

0.895

0.823

0.838

0.515

0.915

 Standard Deviation

0.047

0.074

0.095

0.080

0.092

0.034

CNN3

 Mean

0.921

0.914

0.845

0.857

0.516

0.918

 Standard Deviation

0.035

0.046

0.067

0.059

0.086

0.031