Skip to main content

Table 2 Quantitative results of exploring the effect of preprocessing methods on the segmentation results

From: Lesion segmentation in breast ultrasound images using the optimized marked watershed method

Methods Dice (%)
CLAHE CF SWF Separate(A) Separate(B) Overall
\(\surd \) \(\surd \) \(\surd \) \(\varvec{87.34}\) 86.84 87.09
\(\surd \) \(\surd \) \(\times \) 86.18 87.13 86.66
\(\surd \) \(\times \) \(\surd \) 87.22 \(\varvec{87.52}\) \(\varvec{87.37}\)
\(\times \) \(\times \) \(\surd \) 86.11 86.45 86.28
\(\times \) \(\times \) \(\times \) 77.26 75.80 76.53
  1. Bold indicates the best results in the current column
  2. CLAHE: contrast limited adaptive histogram equalization; CF:curvature filter; SWF: side window filter