MSeg-Net: A Melanoma Mole Segmentation Network Using CornerNet and Fuzzy -Means Clustering

Table 5

Performance analysis of the proposed work with the new methods over the ISIC-2017 dataset.

Reference	Sensitivity	Specificity	Accuracy (%)	Jaccard index (%)	Dice (%)
[71]	0.8804	0.9659	94.50	80.53	87.92
[72]	0.8392	0.9725	93.26	76.53	85.00
[74]	0.9486	0.9889	97.94	—	93.22
[73]	0.9695	0.9950	98.67	95.98	97.95
[75]	0.8600	—	—	—	—
[76]	—	—	94.08	78.55	86.48
[77]	—	—	94.30	—	—
[78]	0.8364	—	93.13	74.88	85.63
Proposed	0.9876	0.9968	99.32	96.93	98.13