Skip to main content

Table 1 Quantitative evaluation of segmentation results for the five methods

From: A shape-optimized framework for kidney segmentation in ultrasound images using NLTV denoising and DRLSE

Method SN SP PPV
DRLSE 0.97±0.02 0.88±0.02 0.78±0.08
DRLSE with NLTV 0.99±0.01 0.94±0.02 0.89±0.04
Bression’s method 0.81±0.10 0.91±0.06 0.80±0.07
Bression’s method with NLTV 0.87±0.04 0.93±0.05 0.87±0.07
Our method 0.96±0.02 0.95±0.02 0.91±0.04
  1. The average and standard deviation of the SN, SP and PPV across the kidney US images are provided.