ZeroReg3D: a zero-shot registration pipeline for 3D consecutive histopathology image reconstruction
We introduced ZeroReg3D, a zero-shot registration pipeline tailored for accurate 3D reconstruction from serial histological sections. By combining zero-shot deep learning-based keypoint matching with optimization-based affine and non-rigid registration techniques, ZeroReg3D effectively addresses critical challenges such as tissue deformation, sectioning artifacts, staining variability, and inconsistent illumination without requiring retraining or fine-tuning.
Authors
- Ruining Deng (ORCID: https://orcid.org/0000-0001-6300-8518)
- Tianyuan Yao (ORCID: https://orcid.org/0000-0002-1848-079X)
- Yuankai Huo (ORCID: https://orcid.org/0000-0002-2096-8065)
- Siqi Lu
- Junchao Zhu (ORCID: https://orcid.org/0000-0002-2576-5312)
- Junlin Guo
- Yuechen Yang
- Chongyu Qu (ORCID: https://orcid.org/0009-0008-8893-2748)
- Mengmeng Yin
- Can Cui (ORCID: https://orcid.org/0000-0002-9517-7823)
- Juming Xiong
- Marilyn Lionts (ORCID: https://orcid.org/0000-0002-9889-3390)
- Jialin Yue
- Haichun Yang
Institutions
- Vanderbilt University (US)
- Cornell University (US)
Publication Details
- Journal
- Journal of Medical Imaging
- Published
- 2025-08-05
- DOI
- https://doi.org/10.1117/1.jmi.12.4.044002
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Science Foundation
- Leona M. and Harry B. Helmsley Charitable Trust
- National Institutes of Health