Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Skeleton Supervised Airway Segmentation

Mingyue Zhao, Han Li, Li Fan, Shiyuan Liu, Xiaolan Qiu, S. Kevin Zhou

TL;DR

A novel skeleton-level annotation (SkA) is introduced, which simplifies the annotation workflow while enhancing annotation consistency and accuracy, preserving the complete topology, and a skeleton-supervised learning framework is proposed to achieve accurate airway segmentation.

Abstract

Fully-supervised airway segmentation has accomplished significant triumphs over the years in aiding pre-operative diagnosis and intra-operative navigation. However, full voxel-level annotation constitutes a labor-intensive and time-consuming task, often plagued by issues such as missing branches, branch annotation discontinuity, or erroneous edge delineation. label-efficient solutions for airway extraction are rarely explored yet primarily demanding in medical practice. To this end, we introduce a novel skeleton-level annotation (SkA) tailored to the airway, which simplifies the annotation workflow while enhancing annotation consistency and accuracy, preserving the complete topology. Furthermore, we propose a skeleton-supervised learning framework to achieve accurate airway segmentation. Firstly, a dual-stream buffer inference is introduced to realize initial label propagation from SkA, avoiding the collapse of direct learning from SkA. Then, we construct a geometry-aware dual-path propagation framework (GDP) to further promote complementary propagation learning, composed of hard geometry-aware propagation learning and soft geometry-aware propagation guidance. Experiments reveal that our proposed framework outperforms the competing methods with SKA, which amounts to only 1.96% airways, and achieves comparable performance with the baseline model that is fully supervised with 100% airways, demonstrating its significant potential in achieving label-efficient segmentation for other tubular structures, such as vessels.

Skeleton Supervised Airway Segmentation

TL;DR

A novel skeleton-level annotation (SkA) is introduced, which simplifies the annotation workflow while enhancing annotation consistency and accuracy, preserving the complete topology, and a skeleton-supervised learning framework is proposed to achieve accurate airway segmentation.

Abstract

Fully-supervised airway segmentation has accomplished significant triumphs over the years in aiding pre-operative diagnosis and intra-operative navigation. However, full voxel-level annotation constitutes a labor-intensive and time-consuming task, often plagued by issues such as missing branches, branch annotation discontinuity, or erroneous edge delineation. label-efficient solutions for airway extraction are rarely explored yet primarily demanding in medical practice. To this end, we introduce a novel skeleton-level annotation (SkA) tailored to the airway, which simplifies the annotation workflow while enhancing annotation consistency and accuracy, preserving the complete topology. Furthermore, we propose a skeleton-supervised learning framework to achieve accurate airway segmentation. Firstly, a dual-stream buffer inference is introduced to realize initial label propagation from SkA, avoiding the collapse of direct learning from SkA. Then, we construct a geometry-aware dual-path propagation framework (GDP) to further promote complementary propagation learning, composed of hard geometry-aware propagation learning and soft geometry-aware propagation guidance. Experiments reveal that our proposed framework outperforms the competing methods with SKA, which amounts to only 1.96% airways, and achieves comparable performance with the baseline model that is fully supervised with 100% airways, demonstrating its significant potential in achieving label-efficient segmentation for other tubular structures, such as vessels.
Paper Structure (4 sections, 8 equations, 4 figures, 2 tables)

This paper contains 4 sections, 8 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Experiment and Results
  4. Conclusion

Figures (4)

  • Figure 1: Motivation for introducing the skeleton level annotation for the airway.
  • Figure 2: Overview of our skeleton-supervised airway segmentation framework. $\mathcal{MP}_g$, $\mathcal{MP}_e$ and $\mathcal{MP}_\odot$ are the mask proposals after G$^{2}$BI, EBI, and DBI, where red, green denotes foreground and unknown regions, the rest area is the background.
  • Figure 3: Visualization of distance maps. (d) and (e), as the reversed maps of (b) and (c), are used for better visual contrast. The $\mathcal{D}_{iggd}$ (f) is much clearer than (d) and (e).
  • Figure 4: Segmentation results on a moderate case (a) and a challenging case (b). Orange boxes and arrows highlight the branches with significant segmentation discrepancies.