Robust Curve Detection in Volumetric Medical Imaging via Attraction Field

TL;DR

This paper presents a neural network-based method for detecting non-branching curves in volumetric medical images, achieving high accuracy without prior shape or orientation knowledge, and provides a new annotated dataset for benchmarking.

Contribution

Introduces a novel neural network approach for robust curve detection in medical imaging that does not rely on domain-specific features or prior object knowledge.

Findings

Achieves subpixel accuracy in curve detection

Outperforms existing methods on diverse clinical tasks

Provides a new annotated dataset for benchmarking

Abstract

Understanding body part geometry is crucial for precise medical diagnostics. Curves effectively describe anatomical structures and are widely used in medical imaging applications related to cardiovascular, respiratory, and skeletal diseases. Traditional curve detection methods are often task-specific, relying heavily on domain-specific features, limiting their broader applicability. This paper introduces a novel approach for detecting non-branching curves, which does not require prior knowledge of the object's orientation, shape, or position. Our method uses neural networks to predict (1) an attraction field, which offers subpixel accuracy, and (2) a closeness map, which limits the region of interest and essentially eliminates outliers far from the desired curve. We tested our curve detector on several clinically relevant tasks with diverse morphologies and achieved impressive subpixel-level accuracy results that surpass existing methods, highlighting its versatility and robustness. Additionally, to support further advancements in this field, we provide our private annotations of aortic centerlines and masks, which can serve as a benchmark for future research. The dataset can be found at https://github.com/neuro-ml/curve-detection.