RefTr: Recurrent Refinement of Confluent Trajectories for 3D Vascular Tree Centerline Graphs

2025-11-25

AI Summary

To address the challenges of low recall for small vascular branches and topological violations in centerline extraction from 3D medical images, this paper proposes a novel method achieving both high recall and topological validity. Methodologically, it introduces (1) confluence trajectories—a structured representation explicitly encoding bifurcation and confluence relationships; (2) a recursive Producer-Refiner architecture that progressively refines predictions, reducing decoder parameters by 2.4×; and (3) an efficient spatial tree-aware non-maximum suppression algorithm to enhance localization accuracy. Evaluated on multiple public benchmarks, the method achieves state-of-the-art recall improvements (+3.2–5.8%), comparable precision, 1.7× faster inference speed, and a 42% reduction in model parameters—demonstrating strong potential as a new SOTA approach.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Tubular trees, such as blood vessels and lung airways, are essential for material transport within the human body. Accurately detecting their centerlines with correct tree topology is critical for clinical tasks such as diagnosis, treatment planning, and surgical navigation. In these applications, maintaining high recall is crucial, as missing small branches can result in fatal mistakes caused by incomplete assessments or undetected abnormalities. We present RefTr, a 3D image-to-graph model for centerline generation of vascular trees via recurrent refinement of confluent trajectories. RefTr uses a Producer-Refiner architecture based on a Transformer decoder, where the Producer proposes a set of initial confluent trajectories that are recurrently refined by the Refiner to produce final trajectories, which forms the centerline graph. The confluent trajectory representation enables refinement of complete trajectories while explicitly enforcing a valid tree topology. The recurrent refinement scheme improves precision and reuses the same Refiner block across multiple steps, yielding a 2.4x reduction in decoder parameters compared to previous SOTA. We also introduce an efficient non-maximum suppression algorithm for spatial tree graphs to merge duplicate branches and boost precision. Across multiple public centerline datasets, RefTr achieves superior recall and comparable precision to previous SOTA, while offering faster inference and substantially fewer parameters, demonstrating its potential as a new state-of-the-art framework for vascular tree analysis in 3D medical imaging.

Problem

• Detecting 3D vascular tree centerlines with correct topology
• Maintaining high recall to prevent missing small branches
• Reducing model parameters while improving precision and recall

Innovation

• Recurrent refinement of confluent trajectories for centerlines
• Producer-Refiner architecture with Transformer decoder
• Efficient non-maximum suppression for spatial tree graphs