Abstract:The rapid and precise segmentation of cell nuclei from hematoxylin and eosin-stained tissue images is an essential clinical undertaking with significant implications for various clinical applications. The segmentation of cell nuclei poses specific challenges due to the inherent instability of nuclear morphology and the complexity of the segmentation environments. Furthermore, previous studies have primarily relied on small-scale and limited-diverse datasets, potentially hindering their applicability to clinical tasks. This study introduces a novel approach, the Double-stage Codec Attention Network, designed to automatically and accurately segment nuclei. Specifically, we present a hierarchical feature extraction module, which maximizes the utilization of cell nuclei's morphological characteristics in the tissue, thereby providing critical semantic information for nucleus segmentation. Furthermore, the feature selection units are employed to enhance relevant features and suppress interfering ones, thereby enhancing the overall expressive capacity of the information. The multi-scale deep feature fusion module utilizes interrelated encoder–decoder connections to jointly optimize and integrate this information, generating a robust hierarchical feature pyramid. Finally, the feature attention fusion mechanism captures spatial and directional information, aiding the model in the accurate localization and recognition of cell nuclei. We rigorously evaluated our proposed method using the PanNuke dataset, the largest comprehensive histology dataset of cancer tissues. In terms of the average F1-score across all segmentation classes in the PanNuke dataset, DSCA-Net outperforms other state-of-the-art models such as DeepLabV3+, TransUNet, Triple U-net, and TransNuSeg by 1.38, 1.44, 2.64, and 1.02, respectively. Additionally, DSCA-Net shows excellent efficiency in generating predictive images, outperforming all comparative models.

Classes U-Net: A method for nuclei segmentation of photoacoustic histology imaging based on information entropy image classification

Mutual-Complementing Framework for Nuclei Detection and Segmentation in Pathology Image

Improved Segmentation of Cellular Nuclei Using UNET Architectures for Enhanced Pathology Imaging

MIU-Net: MIX-Attention and Inception U-Net for Histopathology Image Nuclei Segmentation

DSCA-Net: Double-stage Codec Attention Network for automatic nuclear segmentation

FRE-Net: Full-region enhanced network for nuclei segmentation in histopathology images

Hybrid deep learning network for vascular segmentation in photoacoustic imaging

High-resolution deep transferred ASPPU-Net for nuclei segmentation of histopathology images

Multimodal brain tumor image segmentation based on DenseNet

Multifunctional aggregation network of cell nuclei segmentation aiming histopathological diagnosis assistance: A new MA-Net construction

RIC-Unet: An Improved Neural Network Based on Unet for Nuclei Segmentation in Histology Images

SFE-Net: Spatial-frequency Enhancement Network for Robust Nuclei Segmentation in Histopathology Images

A dual decoder U-Net-based model for nuclei instance segmentation in hematoxylin and eosin-stained histological images

Dual U-Net for the Segmentation of Overlapping Glioma Nuclei.

A Hybrid-Attention Nested UNet for Nuclear Segmentation in Histopathological Images

W-Net: A Two-Stage Convolutional Network for Nucleus Detection in Histopathology Image

Label-Free Nuclei Segmentation Using Intra-Image Self Similarity

Cervical Nuclei Segmentation in Whole Slide Histopathology Images Using Convolution Neural Network

Breast Tumor Tissue Image Classification Using DIU-Net

Nuclear Segmentation in Histopathological Images Using Two-Stage Stacked U-Nets With Attention Mechanism

HoVer-Net: Simultaneous Segmentation and Classification of Nuclei in Multi-Tissue Histology Images