Table 6 Performance comparison between the proposed segmentation and other state-of-the-art methods on ISBI 2017 challenge test set
From: DermoNet: densely linked convolutional neural network for efficient skin lesion segmentation
Method | JC (%) |
|---|---|
FCN-ensemble[18] | 76.5 |
Modified U-Net[33] | 76.2 |
F C N - ( J C ) t r a i n e d w i t h D e r m o S a f e d a t a | 66.49 |
U - N e t - ( J C ) t r a i n e d w i t h D e r m o S a f e d a t a | 75.50 |
D e r m o N e t ( J C ) t r a i n e d w i t h o u t D e r m o S a f e d a t a | 77.5 |
D e r m o N e t ( J C ) t r a i n e d w i t h D e r m o S a f e d a t a | 78.3 |