Collected molecules will appear here. Add from search or explore.
Enhances the Segment Anything Model (SAM) for few-shot medical image segmentation using prototype-guided prompt learning to improve efficiency and accuracy with limited data.
stars
2
forks
1
This is an academic research repository associated with an ISBI 2025 paper. While it addresses a specific niche (few-shot medical segmentation), it has minimal stars and adoption, functioning primarily as a code release for peer review. Its defensibility is low as it's a specific architectural tweak on top of Meta's SAM, which could be superseded by better foundation models or similar research-level prompt engineering techniques.
TECH STACK
INTEGRATION
reference_implementation
READINESS