Residual motion compensation in ECG-gated interventional cardiac vasculature reconstruction

Abstract

Abstract. 3-D reconstruction of cardiac vasculature from angiographic C-arm CT (rotational angiography) data is a major challenge. Motion artefacts corrupt image quality, reducing usability for diagnosis and guidance. Many state-of-theart approaches depend on retrospective ECG-gating of projection data for image reconstruction. A trade-off has to be made regarding the size of the ECG-gating window. A large temporal window is desirable to avoid undersampling. However, residual motion will occur in a large window, causing motion artefacts.We present an algorithm to correct for residual motion. Our approach is based on a deformable 2-D-2-D registration between the forward projection of an initial, ECG-gated reconstruction, and the original projection data. The approach is fully automatic and does not require any complex segmentation of vasculature, or landmarks. The estimated motion is compensated for during the backprojection step of a subsequent reconstruction. We evaluated the method using the publicly available cavarev platform and on six human clinical datasets. We found a better visibility of structure, reduced motion artefacts, and increased sharpness of the vessels in the compensated reconstructions compared to the initial reconstructions. At the time of writing, our algorithm outperforms the leading result of the cavarev ranking list. For the clinical datasets, we found an average reduction of motion artefacts by 13±6%. Vessel sharpness was improved by 25±12% on average.

Reconstruction of coronary arteries from X-ray angiography: A review

“…The shape of the windowing function introduces a weighting to X-ray images depending on temporal distance of X-ray image to the selected phase. Most commonly used windowing functions are nearest-neighbour (Schäfer et al, 2006;Rasche et al, 2006b) or power of cosine function (Schäfer et al, 2006;Rohkohl et al, 2008a;Schwemmer et al, 2013b). The nearest-neighbour function selects the image that is closest to the selected cardiac phase.…”

“…A good suppression can be achieved, however the size of the filter may effect the results (Liu et al, 2014a). On the other hand, some segmentation based background removal algorithms are proposed (Blondel et al, 2006;Zhou et al, 2008;Hu et al, 2010Hu et al, , 2012Schwemmer et al, 2013b). These methods first segment the coronary arteries from X-ray images and remove corresponding pixels.…”

“…These pixels are filled with intensity value estimates from neighbouring background structures to obtain a background image. Hysteresis thresholding of tubularity image (Blondel et al, 2006;Zhou et al, 2008), thresholding of top-hat filtered image (Hu et al, 2010;Schwemmer et al, 2013b), and level set based segmentation are employed. To fill the removed pixels, morphological closure (Blondel et al, 2006;Zhou et al, 2008) and image inpainting are used.…”

“…These attenuation value metrics are reasonable indicators of the contrast and artifact in the resulting reconstruction. Another way of assessing the image artifacts is to employ a noise estimator (Schwemmer et al, 2013b). Another popular choice for assessment is to compare a thresholded reconstruction with the binary ground truth.…”

“…For this purpose, sharpness metric is adapted in (Schwemmer et al, 2013b(Schwemmer et al, , 2014. The centreline is semi-automatically extracted from the reconstructed image and intensity profiles perpendicular to the centrelines are computed.…”