Eso-Net: A Novel 2.5D Segmentation Network With the Multi-Structure Response Filter for the Cancerous Esophagus

Donghao Zhou, Guoheng Huang, Jiajian Li, Siyu Zhu, Zhuowei Wang, Bingo Wing‐Kuen Ling, Chi‐Man Pun, Lianglun Cheng, Xiuyu Cai, Jian ZhouIEEE Access 2020 · DOI

Introduction

Automatic segmentation of the cancerous esophagus in computed tomography (CT) images is a critical computer-assisted method for improving the efficiency of diagnosis and treatment planning, particularly in radiotherapy. However, this task is notoriously difficult due to the high anatomical variability of the esophagus across different cancer stages and locations. Furthermore, the low contrast between the esophagus and surrounding soft tissues often results in "blurry" boundaries that are nearly indistinguishable to standard segmentation algorithms.

Existing approaches typically fall into two categories: 2D networks that ignore vital depth (z-axis) information, and 3D networks that, while spatially aware, often suffer from high computational costs and overfitting on limited medical datasets. To bridge this gap, we present Eso-Net, a 2.5D segmentation network designed to leverage 3D structural priors while maintaining the efficiency of 2D processing.

The Eso-Net Architecture

Eso-Net is built on an encoder-decoder architecture and utilizes a 2.5D approach, taking three consecutive CT slices as input to segment the middle slice. This configuration captures local volumetric context without the memory overhead of full 3D models.

Multi-Structure Response Filter (MSRF)

A key innovation in Eso-Net is the Multi-Structure Response Filter (MSRF). Instead of attempting to enhance the irregular cancerous tissue directly, the MSRF uses Hessian-based 3D filters to identify and enhance regular surrounding structures, such as tube-like blood vessels or plate-like membranes. By defining the "normal" geometry of the background, the network can more effectively isolate the "abnormal" and irregular cancerous regions.

Examples of esophageal cancer segmentation across different CT slices, demonstrating anatomical variability and the model's ability to handle low-contrast boundaries.

Prior Attention Module (PAM)

To integrate these geometric priors, we developed the Prior Attention Module (PAM). Unlike standard self-attention mechanisms that learn spatial relevance from scratch, PAM uses the MSRF output as an explicit spatial prior. This guides the network's attention toward relevant regions during feature transmission through skip connections, significantly improving boundary precision.

Experimental Results

The model was evaluated on a dataset of 30 esophageal cancer patients (6,362 slices) from the Sun Yat-sen University Cancer Center. Using five-fold cross-validation, Eso-Net achieved a Dice Similarity Coefficient (DSC) of 84.839% and a Hausdorff Distance (HD) of 2.583 mm.

A detailed view of a segmented cancerous esophagus (red) within a thoracic CT slice, illustrating the challenge of low-contrast boundaries.

As shown in the figures, Eso-Net demonstrates a superior ability to distinguish the esophagus from neighboring tissues that share similar intensities. Quantitatively, it outperformed established benchmarks including U-Net, 3D U-Net, and Attention U-Net across all key metrics.

Conclusion

Eso-Net addresses the primary challenges of esophageal cancer segmentation by combining 2.5D efficiency with 3D structural priors. The "inverse" enhancement strategy of the MSRF provides a robust way to handle irregular tumor shapes. Future work will focus on making the MSRF parameters learnable to improve generalization across different imaging modalities and organ types.

Summary updated

Abstract

Automatic segmentation of the cancerous esophagus in computed tomography (CT) images is a computer-assisted method that can improve the efficiency of the diagnosis and treatment. Due to the diversity of the cancer stage and location, the anatomical structure of the cancerous esophagus is various. Moreover, the low contrast against surrounding tissues leads to a blurry boundary of the cancerous esophagus. Therefore, existing segmentation networks cannot achieve satisfactory results in automatic segmentation of the cancerous esophagus. In this article, we propose a novel 2.5D segmentation network named Eso-Net for the cancerous esophagus based on an encoder-decoder architecture. A 3D enhancement filter called Multi-Structure Response Filter (MSRF) is designed to extract 3D structural information as prior knowledge. Furthermore, dilated convolutions and residual connections are employed in the convolutional blocks of Eso-Net for multi-scale feature learning. With 3D structural priors, Prior Attention Modules (PAM) are incorporated into the network to facilitate the transmission of relevant spatial information. The experiments are conducted on the dataset from 30 esophageal cancer patients, and we report an 84.839% dice similarity coefficient, an 85.955% precision, an 83.752% sensitivity, and a 2.583mm Hausdorff distance. The experimental results demonstrate that the proposed method outperforms other existing segmentation networks in this task and can effectively assist doctors in the diagnosis and treatment of esophageal cancer.
BibTeX
@inproceedings{zhou2020eso,
  author = {Donghao Zhou and Guoheng Huang and Jiajian Li and Siyu Zhu and Zhuowei Wang and Bingo Wing‐Kuen Ling and Chi‐Man Pun and Lianglun Cheng and Xiuyu Cai and Jian Zhou},
  title = {{Eso-Net: A Novel 2.5D Segmentation Network With the Multi-Structure Response Filter for the Cancerous Esophagus}},
  booktitle = {IEEE Access},
  year = {2020},
  doi = {10.1109/access.2020.3019518},
}
Figure