What Makes for Automatic Reconstruction of Pulmonary Segments

TL;DR

This paper introduces ImPulSe, a deep implicit surface model for automatic pulmonary segment reconstruction, emphasizing anatomical accuracy, efficiency, and the importance of input choices, advancing surgical planning tools.

Contribution

The paper formulates anatomical definitions, proposes evaluation metrics, and introduces ImPulSe, a novel deep implicit surface model for pulmonary segment reconstruction with improved accuracy and efficiency.

Findings

ImPulSe achieves accurate and visually appealing pulmonary segment reconstructions.

Compared to canonical methods, ImPulSe offers higher resolution predictions with fewer parameters.

Analysis of network inputs reveals key factors influencing reconstruction quality.

Abstract

3D reconstruction of pulmonary segments plays an important role in surgical treatment planning of lung cancer, which facilitates preservation of pulmonary function and helps ensure low recurrence rates. However, automatic reconstruction of pulmonary segments remains unexplored in the era of deep learning. In this paper, we investigate what makes for automatic reconstruction of pulmonary segments. First and foremost, we formulate, clinically and geometrically, the anatomical definitions of pulmonary segments, and propose evaluation metrics adhering to these definitions. Second, we propose ImPulSe (Implicit Pulmonary Segment), a deep implicit surface model designed for pulmonary segment reconstruction. The automatic reconstruction of pulmonary segments by ImPulSe is accurate in metrics and visually appealing. Compared with canonical segmentation methods, ImPulSe outputs continuous predictions of arbitrary resolutions with higher training efficiency and fewer parameters. Lastly, we experiment with different network inputs to analyze what matters in the task of pulmonary segment reconstruction. Our code is available at https://github.com/M3DV/ImPulSe.