BREA-Depth: Bronchoscopy Realistic Airway-geometric Depth Estimation

2025-09-15

AI Summary

Monocular bronchoscopic depth estimation suffers from local texture interference and insufficient anatomical awareness in complex, branching airways—especially under low illumination and weak depth cues—leading to navigation inaccuracies and increased surgical risk. To address this, we propose a geometry-prior-guided depth-aware CycleGAN framework. Our method introduces an airway-structure-aware loss that incorporates anatomical priors—including airway centerlines and branch topology—to enforce structural consistency in predicted depth maps. We further formulate a joint optimization objective combining intraluminal depth consistency and 3D structural integrity. Additionally, we introduce the first airway-specific depth-structure evaluation metric. Evaluated on ex vivo human lung data and public bronchoscopy datasets, our approach outperforms state-of-the-art methods, achieving a 21.3% reduction in depth error and significantly improved robustness of 3D reconstruction and anatomical fidelity.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Structured Learning
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Monocular depth estimation in bronchoscopy can significantly improve real-time navigation accuracy and enhance the safety of interventions in complex, branching airways. Recent advances in depth foundation models have shown promise for endoscopic scenarios, yet these models often lack anatomical awareness in bronchoscopy, overfitting to local textures rather than capturing the global airway structure, particularly under ambiguous depth cues and poor lighting. To address this, we propose Brea-Depth, a novel framework that integrates airway-specific geometric priors into foundation model adaptation for bronchoscopic depth estimation. Our method introduces a depth-aware CycleGAN, refining the translation between real bronchoscopic images and airway geometries from anatomical data, effectively bridging the domain gap. In addition, we introduce an airway structure awareness loss to enforce depth consistency within the airway lumen while preserving smooth transitions and structural integrity. By incorporating anatomical priors, Brea-Depth enhances model generalization and yields more robust, accurate 3D airway reconstructions. To assess anatomical realism, we introduce Airway Depth Structure Evaluation, a new metric for structural consistency. We validate BREA-Depth on a collected ex vivo human lung dataset and an open bronchoscopic dataset, where it outperforms existing methods in anatomical depth preservation.

Problem

• Monocular depth estimation in bronchoscopy for navigation accuracy
• Addressing anatomical unawareness in endoscopic depth models
• Integrating airway-specific geometric priors into foundation models

Innovation

• Integrates airway-specific geometric priors
• Uses depth-aware CycleGAN for domain translation
• Introduces airway structure awareness loss