PRDL learns a promptable Gaussian distribution over patch representations during DINO-style pretraining and uses it to augment WSI classifiers, improving AUC on lung EGFR and cancer subtyping benchmarks.
This strategy generates 457,000, 422,000, and 1,839,000 patches, respectively for USTC-EGFR, TCGA- EGFR, and TCGA-LUNG-3K dataset, for the training of the self-supervised models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Promptable Representation Distribution Learning and Data Augmentation for Gigapixel Histopathology WSI Analysis
PRDL learns a promptable Gaussian distribution over patch representations during DINO-style pretraining and uses it to augment WSI classifiers, improving AUC on lung EGFR and cancer subtyping benchmarks.