rictoo/hoverlink-puma: HoverLink v1.0: PUMA Challenge Submission
Abstract
Official submission for the Panoptic Segmentation of Nuclei and Tissue in Advanced Melanoma (PUMA) Challenge. 🏆 PUMA Track 1 Results: 3rd Place Overall 1st Place Nuclear Segmentation 📦 Contents: Pipeline: End-to-end training (training.py) and inference (inference.py). Model: Dual-branch U-Net (ConvNeXt-Base) + feature linkage. Utils: Data preprocessing (preprocessing/) and hyperparam search (hp_search.ipynb). ⚖️ License: See README.md for full details.
Fields of Study & Subjects
AI in cancer detection
Cell Image Analysis Techniques
Medical Image Segmentation Techniques
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 9 February 2026