The authors formulate mammographic lesion segmentation as a mixed-supervision open-vocabulary problem.
Mixed-supervision method improves mammography segmentation for rare lesions
The authors combine pixel-level masks with image-level labels to train a segmentation model that generalizes to lesion categories lacking dense annotations.
Top University
Di Dai · Bo Liu · Liwei Wang
Peking University
Research Digest··3 min read
Dai et al.
Why this paper
From Peking University · Released code
In one line
A two-branch framework with lesion-wise attentive classifier improves mammography segmentation under mixed supervision.
What it released
Code
What we could check
- ✓Code link in the paper (github.com)
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§