Skip to main content

Table 6 Statistical average results for test images of 1 to 20 from the datasets

From: Structure-based level set method for automatic retinal vasculature segmentation

Dataset Method Se Sp Ppv Npv Acc κ
DRIVE Unsupervised PBLS[24] 0.7754 0.9348 0.6403 0.9655 0.9140 0.6494
Jiang et al.[12] - - - - 0.9212 -
Martinez-Perez et al.[13] 0.7246 0.9655 - - 0.9344 -
Proposed 0.7704 0.9613 0.7460 0.9658 0.9365 0.7198
Budai et al.[17] 0.6440 0.9870 - - 0.9572 -
Supervised Staal et al.[16] 0.7194 0.9773 - - 0.9442 -
Marin et al.[14] 0.7067 0.9801 0.8433 0.9582 0.9452 -
Soares et al.[15] 0.7283 0.9788 - - 0.9466 -
Ours Unsupervised PBLS[24] 0.6600 0.9482 0.4380 0.9804 0.9328 0.4754
Proposed 0.5179 0.9810 0.6042 0.9737 0.9567 0.5192
STARE Unsupervised PBLS[24] 0.8268 0.9117 0.5227 0.9803 0.9035 0.5822
Hoover et al.[11] 0.6751 0.9567 - - 0.9267 -
Martinez-Perez et al.[13] 0.7506 0.9569 - - 0.9410 -
Proposed 0.6926 0.9726 0.7633 0.9656 0.9441 0.6779
Supervised Soares et al.[15] 0.7103 0.9737 - - 0.9480 -
Staal et al.[16] 0.6970 0.9810 - - 0.9516 -
   Marin et al.[14] 0.6944 0.9819 - - 0.9526 -