Saved in:
Bibliographic Details
Main Authors: Bai, Lin, Li, Xiaoyang, Huang, Liqiang, Nguyen, Quynh, Van Nguyen, Hien, Prasad, Saurabh, Maric, Dragan, Redell, John, Dash, Pramod, Roysam, Badrinath
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.11722
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918246740918272
author Bai, Lin
Li, Xiaoyang
Huang, Liqiang
Nguyen, Quynh
Van Nguyen, Hien
Prasad, Saurabh
Maric, Dragan
Redell, John
Dash, Pramod
Roysam, Badrinath
author_facet Bai, Lin
Li, Xiaoyang
Huang, Liqiang
Nguyen, Quynh
Van Nguyen, Hien
Prasad, Saurabh
Maric, Dragan
Redell, John
Dash, Pramod
Roysam, Badrinath
contents We present a weak to strong generalization methodology for fully automated training of a multi-head extension of the Mask-RCNN method with efficient channel attention for reliable segmentation of overlapping cell nuclei in multiplex cyclic immunofluorescent (IF) whole-slide images (WSI), and present evidence for pseudo-label correction and coverage expansion, the key phenomena underlying weak to strong generalization. This method can learn to segment de novo a new class of images from a new instrument and/or a new imaging protocol without the need for human annotations. We also present metrics for automated self-diagnosis of segmentation quality in production environments, where human visual proofreading of massive WSI images is unaffordable. Our method was benchmarked against five current widely used methods and showed a significant improvement. The code, sample WSI images, and high-resolution segmentation results are provided in open form for community adoption and adaptation.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11722
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Weak-to-Strong Generalization Enables Fully Automated De Novo Training of Multi-head Mask-RCNN Model for Segmenting Densely Overlapping Cell Nuclei in Multiplex Whole-slice Brain Images
Bai, Lin
Li, Xiaoyang
Huang, Liqiang
Nguyen, Quynh
Van Nguyen, Hien
Prasad, Saurabh
Maric, Dragan
Redell, John
Dash, Pramod
Roysam, Badrinath
Computer Vision and Pattern Recognition
We present a weak to strong generalization methodology for fully automated training of a multi-head extension of the Mask-RCNN method with efficient channel attention for reliable segmentation of overlapping cell nuclei in multiplex cyclic immunofluorescent (IF) whole-slide images (WSI), and present evidence for pseudo-label correction and coverage expansion, the key phenomena underlying weak to strong generalization. This method can learn to segment de novo a new class of images from a new instrument and/or a new imaging protocol without the need for human annotations. We also present metrics for automated self-diagnosis of segmentation quality in production environments, where human visual proofreading of massive WSI images is unaffordable. Our method was benchmarked against five current widely used methods and showed a significant improvement. The code, sample WSI images, and high-resolution segmentation results are provided in open form for community adoption and adaptation.
title Weak-to-Strong Generalization Enables Fully Automated De Novo Training of Multi-head Mask-RCNN Model for Segmenting Densely Overlapping Cell Nuclei in Multiplex Whole-slice Brain Images
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.11722