Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DigestPath

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-class Image ClassificationDigestPath 2019 (test)
Accuracy94.91
32
Semantic SegmentationDigestPath (test)
DSC77.47
29
Tile-level ClassificationDigestPath
Weighted F1 Score93.5
16
Semantic SegmentationDigestPath 68 WSIs (val)
DSC82.92
14
Colorectal histopathology segmentationDigestPath WSI
Accuracy98.73
11
Colorectal histopathology segmentationDigestPath Patch
Accuracy97.69
11
Histopathology Image ClassificationDigestPath
Accuracy84.84
10
Pathological Detection and SegmentationDigestPath 2019 (test)
DSC80.75
10
WSI-level segmentationDigestPath
Dice85.1
8
Patch-level Histopathology ClassificationDigestPath (5-fold CV)
Macro F1 Score95.82
7
SegmentationDigestPath (internal held-out)
Dice Score88.8
5
Interactive SegmentationDigestpath
Number of Clicks @ 85%16.46
3
Patch-Level ClassificationDigestPath
Accuracy96.73
3
Semantic SegmentationDigestPath Sampled lesion patches (val)
DSC87.49
3
Showing 14 of 14 rows