Overview
Who this competition fits
适合已有医学图像深度学习和眼底彩照(CFP)分割经验的人,尤其是能处理荧光素眼底血管造影(FFA)与CFP跨模态信息的参赛者。也适合能够从血管分割结果进一步实现CRAE、CRVE、AVR、血管密度和分形维数计算的团队;可个人或至多5人组队参加。
Read the original official blurb
MICCAI2026 Challenge: Generalized Analysis of Vessels in Eye Edition 2 (GAVE2) Official tags: CV. Sign-ups: 247.
Resources
What this competition gives you
There is a baseline you can run as-is — starting from it is usually far faster than building from scratch.
- BaselineOfficial GitHub baseline code for the challenge
- Data150 annotated CFP–FFA pairs, with training, validation, and blind final splits
- Reference paperREFUGE challenge review paper
- Reference paperAGE challenge review paper
- Reference paperADAM Challenge paper
- Reference paperREFUGE2 Challenge paper
- Reference paperGamma challenge paper
Preparation
From registration to a first submission
- 01
眼底图像处理
- 02
动静脉语义分割
- 03
CFP与FFA多模态建模
- 04
医学图像深度学习
- 05
血管形态学量化
- 06
结果文件提交与代码复现
Before you commit: 难点不只是把血管分出来,而是要区分动脉和静脉,并处理CFP中“色彩亮度相近”、光照变化和血管交叉边界模糊的问题。若想在综合排名中有竞争力,最好同时完成三个子任务:任务一又明确禁止在训练和测试中使用FFA,而任务二需要利用FFA的动态灌注信息,任务三还要求将分割转化为多项定量血管指标。
Source
How this page was assembled
Competition information is structured from the official page. Scores are platform estimates for decision support; official rules take precedence.
- Official competition page
- Baidu AI Studio
- Last checked
- Not recorded