Adversarial Transformer for Repairing Human Airway Segmentation

TL;DR

This paper introduces an adversarial refinement network that improves airway segmentation accuracy in CT images, addressing discontinuities and heterogeneity issues, with validated results across multiple datasets and models.

Contribution

A novel patch-scale adversarial refinement method that enhances existing airway segmentation models and generalizes across different datasets and architectures.

Findings

Over 15% improvement in detected airway length and branch ratios.

Effective detection of discontinuities and missing bronchioles.

Significant enhancement of segmentation completeness across models.

Abstract

Discontinuity in the delineation of peripheral bronchioles hinders the potential clinical application of automated airway segmentation models. Moreover, the deployment of such models is limited by the data heterogeneity across different centres, and pathological abnormalities also make achieving accurate robust segmentation in distal small airways difficult. Meanwhile, the diagnosis and prognosis of lung diseases often rely on evaluating structural changes in those anatomical regions. To address this gap, this paper presents a patch-scale adversarial-based refinement network that takes in preliminary segmentation along with original CT images and outputs a refined mask of the airway structure. The method is validated on three different datasets encompassing healthy cases, cases with cystic fibrosis and cases with COVID-19. The results are quantitatively evaluated by seven metrics and achieved more than a 15% rise in detected length ratio and detected branch ratio, showing promising performance compared to previously proposed models. The visual illustration also proves our refinement guided by a patch-scale discriminator and centreline objective functions is effective in detecting discontinuities and missing bronchioles. Furthermore, the generalizability of our refinement pipeline is tested on three previous models and improves their segmentation completeness significantly.