Hide-and-Seek Attribution combined with a diffusion autoencoder converts coarse vertebra-level labels into accurate lytic and blastic lesion segmentations, reaching F1 scores of 0.91 and 0.85 without any mask supervision.
Applying multiple instance learning for breast cancer lesion detection in mammography images
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Hide-and-Seek Attribution: Weakly Supervised Segmentation of Vertebral Metastases in CT
Hide-and-Seek Attribution combined with a diffusion autoencoder converts coarse vertebra-level labels into accurate lytic and blastic lesion segmentations, reaching F1 scores of 0.91 and 0.85 without any mask supervision.