Published November 28, 2025 | Version v1
Dataset Open

Segmented Masks for Best-Performing Models in A U-Net-Based Approach for Histological Tissue Segmentation Using RCAug Data Augmentation

Authors/Creators

  • 1. ROR icon Federal Institute of São Paulo
  • 2. ROR icon Universidade Federal de São Paulo
  • 3. Universidade Estadual Paulista Júlio de Mesquita Filho Câmpus de São José do Rio Preto Instituto de Biociências Letras e Ciências Exatas
  • 4. ROR icon Universidade Federal de Uberlândia

Description

This dataset contains the segmentation masks corresponding to the best-performing model and augmentation configurations reported in the paper “A U-Net-Based Approach for Histological Tissue Segmentation Using RCAug Data Augmentation” (SIBGRAPI 2025, IEEE Xplore).

For each histology dataset, we provide the predicted masks generated by the top-performing U-Net-based models under the augmentation strategies highlighted in the article, together with concise summary metrics. These files are intended to support result inspection, comparison and reuse in further studies.

Files

mean-std-dice-by-model-and-augmentation.csv

Files (183.4 MB)

Name Size Download all
md5:016249a0be7a6145670f69192d2e22e1
2.4 kB Preview Download
md5:815d512e6b58c30a45e18bdbf2dd81d0
183.4 MB Preview Download

Additional details

Funding

National Council for Scientific and Technological Development
305386/2024-7
National Council for Scientific and Technological Development
302833/2025-0
Fundação de Amparo à Pesquisa do Estado de Minas Gerais
APQ-00727-24
Fundação de Amparo à Pesquisa do Estado de São Paulo
2022/03020-1

Software

Repository URL
https://github.com/LIPAI-Org/unet-rcaug-histoseg
Programming language
Python
Development Status
Active