# Bending Loss Regularized Network for Nuclei Segmentation in Histopathology Images

arXiv: https://arxiv.org/abs/2002.01020 · HTML: https://mlsanity.com/paper/2002.01020/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Feb 2020.

## Built on by (most cited first, 10 of 10)

- [Panoptic Segmentation: A Review](https://mlsanity.com/paper/2111.10250/) (2021)
- [Which Pixel to Annotate: a Label-Efficient Nuclei Segmentation Framework](https://mlsanity.com/paper/2212.10305/) (2022)
- [Sharp-GAN: Sharpness Loss Regularized GAN for Histopathology Image Synthesis](https://mlsanity.com/paper/2110.14709/) (2021)
- [SRPN: similarity-based region proposal networks for nuclei and cells detection in histology images](https://mlsanity.com/paper/2106.13556/) (2021)
- [SIAN: Style-Guided Instance-Adaptive Normalization for Multi-Organ Histopathology Image Synthesis](https://mlsanity.com/paper/2209.02412/) (2022)
- [Nuclei & Glands Instance Segmentation in Histology Images: A Narrative Review](https://mlsanity.com/paper/2208.12460/) (2022)
- [Enhanced Sharp-GAN For Histopathology Image Synthesis](https://mlsanity.com/paper/2301.10187/) (2023)
- [Bend-Net: Bending Loss Regularized Multitask Learning Network for Nuclei Segmentation in Histopathology Images](https://mlsanity.com/paper/2109.15283/) (2021)
- [Meta Mask Correction for Nuclei Segmentation in Histopathological Image](https://mlsanity.com/paper/2111.12498/) (2021)
- [NuSeC: A Dataset for Nuclei Segmentation in Breast Cancer Histopathology Images](https://mlsanity.com/paper/2507.14272/) (2025)

## Abstract (arXiv)

Separating overlapped nuclei is a major challenge in histopathology image analysis. Recently published approaches have achieved promising overall performance on public datasets; however, their performance in segmenting overlapped nuclei are limited. To address the issue, we propose the bending loss regularized network for nuclei segmentation. The proposed bending loss defines high penalties to contour points with large curvatures, and applies small penalties to contour points with small curvature. Minimizing the bending loss can avoid generating contours that encompass multiple nuclei. The proposed approach is validated on the MoNuSeg dataset using five quantitative metrics. It outperforms six state-of-the-art approaches on the following metrics: Aggregate Jaccard Index, Dice, Recognition Quality, and Pan-optic Quality.