Bending Loss Regularized Network for Nuclei Segmentation in Histopathology Images

  title={Bending Loss Regularized Network for Nuclei Segmentation in Histopathology Images},
  author={Haotian Wang and Min Xian and Aleksandar Vakanski},
  journal={2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI)},
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… 

Figures from this paper

Bend-Net: Bending Loss Regularized Multitask Learning Network for Nuclei Segmentation in Histopathology Images
A novel multitask learning network with a bending loss regularizer to separate overlapped nuclei accurately is proposed and extensive experiments demonstrate that the proposed Bend-Net outperforms eight state-of-the-art approaches.
Sharp-GAN: Sharpness Loss Regularized GAN for Histopathology Image Synthesis
A sharpness loss regularized generative adversarial network is proposed to synthesize realistic histopathology images with clear nuclei contours and enhances the contrast ofuclei contour pixels.
Contextual Mixing Feature Unet for Multi-Organ Nuclei Segmentation
A contextual mixing feature Unet (CMF-Unet), which utilizes two parallel branches, nuclei segmentation branch and boundary extraction branch, and mixes complementary feature maps from two branches to obtain rich and integrated contextual features, is proposed.
Meta Mask Correction for Nuclei Segmentation in Histopathological Image
A fully conventional meta-model is designed that can correct noisy masks using a small amount of clean meta-data and then the corrected masks can be used to supervise the training of the segmentation model.
Improving Mask R-CNN for Nuclei Instance Segmentation in Hematoxylin & Eosin-Stained Histological Images
This work proposes a modified Mask R-CNN-based approach by incorporating distance maps of instances and hematoxylin-stain intensities as extra input channels to the model and explores the impact of three well-known inference strategies, namely test-time augmentation, ensembling, and knowledge transfer through pre-training on the segmentation performance.
REU-Net: Region-enhanced nuclei segmentation network.
RDCNet: Instance Segmentation with a Minimalist Recurrent Residual Network
A minimalist recurrent network called recurrent dilated convolutional network (RDCNet), consisting of a shared stacked diluted convolution (sSDC) layer that iteratively refines its output and thereby generates interpretable intermediate predictions, achieves state-of-the-art on 2 of the 3 datasets.
Master Thesis-Medical Image Analysis using Deep Learning
This Master Thesis provides a summary overview on the use of current deep learning-based object detection methods for the analysis of medical images, in particular from microscopic tissue sections, and aims at making the results reproducible.


SAMS-NET: Stain-aware multi-scale network for instance-based nuclei segmentation in histology images
  • S. Graham, N. Rajpoot
  • Physics
    2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)
  • 2018
A deep multi-scale neural network, with a novel loss function that is sensitive to the Hematoxylin intensity, for precise object-level nuclei segmentation for computational precision medicine (CPM) segmentation challenge dataset as part of MICCAI 2017.
Nuclei segmentation in histopathology images using deep neural networks
This work presents a fully automated workflow to segment nuclei from histopathology image data by using deep neural networks trained from a set of manually annotated images and by processing the posterior probability maps in order to split jointly segmented nuclei.
DCAN: Deep Contour-Aware Networks for Accurate Gland Segmentation
An efficient deep contour-aware network (DCAN) to solve this challenging problem under a unified multi-task learning framework and can be efficient when applied to large-scale histopathological data without resorting to additional steps to generate contours based on low-level cues for post-separating.
A Dataset and a Technique for Generalized Nuclear Segmentation for Computational Pathology
A large publicly accessible data set of hematoxylin and eosin (H&E)-stained tissue images with more than 21000 painstakingly annotated nuclear boundaries is introduced, whose quality was validated by a medical doctor.
BESNet: Boundary-Enhanced Segmentation of Cells in Histopathological Images
A new network structure that encodes input images to feature maps similar to U-net but utilizes two decoding paths that restore the original image resolution, which can be used to improve the quality of the entire cell segmentation achieved in the other decoding path.
Hover-Net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images
CIA-Net: Robust Nuclei Instance Segmentation with Contour-aware Information Aggregation
A novel deep neural network, namely Contour-aware Informative Aggregation Network (CIA-Net) with multi-level information aggregation module between two task-specific decoders that leverages the merit of spatial and texture dependencies between nuclei and contour by bi-directionally aggregating task- specific features.
Segmentation of Clustered Nuclei With Shape Markers and Marking Function
A method to separate clustered nuclei from fluorescence microscopy cellular images, using shape markers and marking function in a watershed-like algorithm, achieving 6%-7% improvement of segmentation accuracies over earlier methods is presented.
U-Net: Convolutional Networks for Biomedical Image Segmentation
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.
Structure-Preserving Color Normalization and Sparse Stain Separation for Histological Images
Stain density correlation with ground truth and preference by pathologists were higher for images normalized using the method when compared to other alternatives, and a computationally faster extension of this technique is proposed for large whole-slide images that selects an appropriate patch sample instead of using the entire image to compute the stain color basis.