ConStruct: Structural Distillation of Foundation Models for Prototype-Based Weakly Supervised Histopathology Segmentation Article Swipe

MAKE IMPACT

Khang Le , Ha Thach , Anh M. Vu , Thuy Thi Bich Vo , Hong-Han Huynh , David J. Yang , Minh Hoang Le , Thanh-Huy Nguyen , Akash Awasthi , Chandra Mohan , Zhu Han , Hien Van Nguyen · YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2512.10316

Weakly supervised semantic segmentation (WSSS) in histopathology relies heavily on classification backbones, yet these models often localize only the most discriminative regions and struggle to capture the full spatial extent of tissue structures. Vision-language models such as CONCH offer rich semantic alignment and morphology-aware representations, while modern segmentation backbones like SegFormer preserve fine-grained spatial cues. However, combining these complementary strengths remains challenging, especially under weak supervision and without dense annotations. We propose a prototype learning framework for WSSS in histopathological images that integrates morphology-aware representations from CONCH, multi-scale structural cues from SegFormer, and text-guided semantic alignment to produce prototypes that are simultaneously semantically discriminative and spatially coherent. To effectively leverage these heterogeneous sources, we introduce text-guided prototype initialization that incorporates pathology descriptions to generate more complete and semantically accurate pseudo-masks. A structural distillation mechanism transfers spatial knowledge from SegFormer to preserve fine-grained morphological patterns and local tissue boundaries during prototype learning. Our approach produces high-quality pseudo masks without pixel-level annotations, improves localization completeness, and enhances semantic consistency across tissue types. Experiments on BCSS-WSSS datasets demonstrate that our prototype learning framework outperforms existing WSSS methods while remaining computationally efficient through frozen foundation model backbones and lightweight trainable adapters.

Type

preprint

Landing Page

http://arxiv.org/abs/2512.10316

PDF

https://arxiv.org/pdf/2512.10316

OA Status

green

OpenAlex ID

https://openalex.org/W4417295371

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4417295371

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2512.10316

Digital Object Identifier

Title

ConStruct: Structural Distillation of Foundation Models for Prototype-Based Weakly Supervised Histopathology Segmentation

Work title

Type

preprint

OpenAlex work type

Publication year

2025

Year of publication

Publication date

2025-12-11

Full publication date if available

Authors

Khang Le, Ha Thach, Anh M. Vu, Thuy Thi Bich Vo, Hong-Han Huynh, David J. Yang, Minh Hoang Le, Thanh-Huy Nguyen, Akash Awasthi, Chandra Mohan, Zhu Han, Hien Van Nguyen

List of authors in order

Landing page

https://arxiv.org/abs/2512.10316

Publisher landing page

PDF URL

https://arxiv.org/pdf/2512.10316

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2512.10316

Direct OA link when available

Cited by

0

Total citation count in OpenAlex

Full payload