Motion estimation of skeletonized angiographic images using elastic registration

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Image Processing

Background:

• Digital subtraction angiography (DSA) is crucial for visualizing arterial structures.
• Accurate estimation of arterial motion is vital for diagnosing vascular diseases and guiding interventions.
• Existing motion estimation techniques face challenges with low contrast and noisy angiographic data.

Purpose of the Study:

• To develop and evaluate a novel elastic registration algorithm for precise arterial motion estimation in digital angiographic image sequences.
• To improve the robustness of motion estimation in challenging imaging conditions, such as low contrast and high noise.

Main Methods:

• A recursive elastic registration algorithm is applied to binary skeleton images of arteries.
• The algorithm integrates an autoregressive (AR) model at the pixel level with a dynamic programming (DP) algorithm.
• A moving average (MA) model is combined with the local AR model to enhance overall skeleton registration accuracy.

Main Results:

• The proposed method successfully estimates arterial motion in both simulated and real digital angiographic image sequences.
• The elastic registration of skeletons demonstrates high performance, particularly in scenarios with low contrast and noisy images.
• The combination of AR and MA models leads to improved registration outcomes.

Conclusions:

• Elastic registration of arterial skeletons provides a robust and accurate method for motion estimation in digital angiography.
• This approach offers significant advantages for analyzing challenging angiographic datasets, enhancing diagnostic capabilities.
• The developed technique holds promise for clinical applications requiring precise assessment of arterial dynamics.