Table 3 A summary of Self-Supervised... | Emerald Publishing

Table 3

A summary of Self-Supervised nuclei segmentation methods.

Ref.	Dataset	Methods	Pre-Processing	Post-Processing
[35]	COCO, BBBC, Kumar, TNBC	Mask RCNN based domain adaptation using pseudo labeling	DARCNN pretrained with source dataset	–
[108]	400 single WBC images split into two datasets of 300 and 100 images	Unsupervised initial segmentation with K-means followed by supervised refinement using SVM classifier	Conversion to HSI color space	–
[85]	MoNuSeg	Attention network for scale identification with segmentation maps as auxiliary outputs	Tile extraction, stain normalization	Opening, closing, distance transform
[1]	MoNuSeg, TNBC	Multi scale representation based Self supervised Learning using U-Net	Cropping, resizing, ResNet-18 pretrained with zoomed in and zoomed out tiles	–
[71]	Kaggle DSB18, BUSIS, ISIC18, BraTS18	Redundancy reduction based Barlow Twins U-Net	U-Net encoder pretrained with Barlow Twins approach (Siamese Net)	–
[55]	BBBC039V1, Kumar, TNBC	Cycle Consistency Panoptic Domain Adaptive Mask RCNN	Normalization, random sample cropping, removal of samples with less than 3 objects, complementing	–
[103]	MoNuSeg	Contrastive Learning using Scalewise Triplet Loss and Count Ranking to pretrain U-Net encoder	Anchor, positive and negative tile sampling	–
[5]	MoNuSeg, CoNSeP	Positive and Negative Patch based Contrastive Learning using FCN		–

Ref.	Dataset	Methods	Pre-Processing	Post-Processing
[35]	COCO, BBBC, Kumar, TNBC	Mask RCNN based domain adaptation using pseudo labeling	DARCNN pretrained with source dataset	–
[108]	400 single WBC images split into two datasets of 300 and 100 images	Unsupervised initial segmentation with K-means followed by supervised refinement using SVM classifier	Conversion to HSI color space	–
[85]	MoNuSeg	Attention network for scale identification with segmentation maps as auxiliary outputs	Tile extraction, stain normalization	Opening, closing, distance transform
[1]	MoNuSeg, TNBC	Multi scale representation based Self supervised Learning using U-Net	Cropping, resizing, ResNet-18 pretrained with zoomed in and zoomed out tiles	–
[71]	Kaggle DSB18, BUSIS, ISIC18, BraTS18	Redundancy reduction based Barlow Twins U-Net	U-Net encoder pretrained with Barlow Twins approach (Siamese Net)	–
[55]	BBBC039V1, Kumar, TNBC	Cycle Consistency Panoptic Domain Adaptive Mask RCNN	Normalization, random sample cropping, removal of samples with less than 3 objects, complementing	–
[103]	MoNuSeg	Contrastive Learning using Scalewise Triplet Loss and Count Ranking to pretrain U-Net encoder	Anchor, positive and negative tile sampling	–
[5]	MoNuSeg, CoNSeP	Positive and Negative Patch based Contrastive Learning using FCN		–