Skip to main content

Table 1 Comparison of RV segmentation results of our method and the state-of-the-art methods tested on MICCAI 2012 database.

From: Fully automatic segmentation of heart chambers in cardiac MRI using deep learning

Method A/SA DM HD(mm) R(EDV) R(ESV) R(EF)
Our Method A 0.81(0.21) 7.79(5.91) 0.99 0.99 0.95
CMIC (Zuluaga et al., 2013) A 0.78(0.23) 10.51(9.17) 0.93 0.93 0.73
NTUST (Wang et al., 2012) SA 0.57(0.33) 28.44(23.57) 0.71 0.78 0.23
SBIA (Ou et al., 2011) SA 0.55(0.32) 23.16(19.86) 0.63 0.69 0.45
BIT-UPM (Maier et al., 2012) SA 0.80(0.19) 11.15(6.62) 0.99 0.97 0.84
GEWU(Nambakhsh et al., 2013) SA 0.59(0.24) 20.21(9.72) 0.81 0.81 0.4
ICL(Bai et al., 2013) SA 0.78(0.2) 9.26(4.93) 0.98 0.98 0.85
LITIS(Grosgeorge et al., 2013) SA 0.76(0.2) 9.97(5.49) 0.95 0.90 0.76
  1. A: Automatic, SA: semi-Automatic. Hausdorff Distance (HD), Dice Metric (DM), correlation coefficient (R).