CLISC: Bridging CLIP and SAM by Enhanced CAM for Unsupervised Brain Tumor Segmentation
Xiaochuan Ma, Jia Fu, Wenjun Liao, Shichuan Zhang, Guotai Wang · 2025
Read CLISC: Bridging CLIP and SAM by Enhanced CAM for Unsupervised Brain Tumor Segmentation by Xiaochuan Ma, Jia Fu, Wenjun Liao, Shichuan Zhang, Guotai Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-16246.txt.
Description
Xiaochuan Ma, Jia Fu, Wenjun Liao, Shichuan Zhang, and Guotai Wang present a way to outline brain tumors in scans without any hand labels. Their paper is from two thousand twenty-five. Labeling scans by hand is slow and costly, yet most deep learning methods need many labeled examples to learn from.
Their method chains two large pre-trained models together. The first flags which scans hold a tumor and marks rough regions of interest. The second turns those rough marks into clean outlines. A third network then learns from those outlines, while poor ones are filtered out. On a standard test set, the method beat five rival label-free methods by more than ten points. It came close to fully supervised results.
Catalog record
| CREATOR | Xiaochuan Ma, Jia Fu, Wenjun Liao, Shichuan Zhang, Guotai Wang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 19 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-16246