Robust Curve Detection in Volumetric Medical Imaging via Attraction Field
This addresses the need for robust and versatile curve detection in medical diagnostics for diseases like cardiovascular, respiratory, and skeletal conditions, though it is incremental as it builds on neural network approaches.
The paper tackled the problem of detecting non-branching curves in volumetric medical imaging without prior knowledge of object orientation, shape, or position, achieving subpixel-level accuracy that surpasses existing methods on clinically relevant tasks.
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.