Weighting the fine-tuning loss by a pixel-level transfer risk map improves brain MRI segmentation Dice by 1.8% to 4.4% over vanilla fine-tuning.
Transfer learning techniques for medical image analysis: A review,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Transfer Risk Map: Mitigating Pixel-level Negative Transfer in Medical Segmentation
Weighting the fine-tuning loss by a pixel-level transfer risk map improves brain MRI segmentation Dice by 1.8% to 4.4% over vanilla fine-tuning.