MIAS-SAM generates medical anomaly segmentations by memory-bank matching of SAM encoder features, using the anomaly map's center of gravity as a point prompt for the SAM decoder, without thresholding the anomaly map.
Autoencoders for unsupervised anomaly segmentation in brain mr images: a comparative study.Medical image analysis, 69:101952, 2021
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
support 1representative citing papers
citing papers explorer
-
MIAS-SAM: Medical Image Anomaly Segmentation without thresholding
MIAS-SAM generates medical anomaly segmentations by memory-bank matching of SAM encoder features, using the anomaly map's center of gravity as a point prompt for the SAM decoder, without thresholding the anomaly map.