[Paper Review] Model Based Iterative Reconstruction With Spatially Adaptive Sinogram Weights for Wide-Cone Cardiac CT

Amirkoushyar Ziabari, Dong Hye Ye+5 more|arXiv (Cornell University)|Dec 20, 2018

Medical Imaging Techniques and Applications14 references4 citations

TL;DR

This paper proposes SAW-MBIR, a model-based iterative reconstruction method with spatially-adaptive sinogram weights that selectively uses half-scan data for the heart region to preserve high temporal resolution and full-scan data for the rest of the volume to ensure stable, artifact-free image quality in wide-cone cardiac CT. The method achieves half-scan temporal resolution performance while matching full-scan image quality outside the heart, validated on clinical whole-heart CT data.

ABSTRACT

With the recent introduction of CT scanners with large cone angles, wide coverage detectors now provide a desirable scanning platform for cardiac CT that allows whole heart imaging in a single rotation. On these scanners, while half-scan data is strictly sufficient to produce images with the best temporal resolution, acquiring a full 360 degree rotation worth of data is beneficial for wide-cone image reconstruction at negligible additional radiation dose. Applying Model-Based Iterative Reconstruction (MBIR) algorithm to the heart has shown to yield significant enhancement in image quality for cardiac CT. But imaging the heart in large cone angle geometry leads to apparently conflicting data usage considerations. On the one hand, in addition to using the fastest available scanner rotation speed, a minimal complete data set of 180 degrees plus the fan angle is typically used to minimize both cardiac and respiratory motion. On the other hand, a full 360 degree acquisition helps better handle the challenges of missing frequencies and incomplete projections associated with wide-cone half-scan data acquisition. In this paper, we develop a Spatially Adaptive sinogram Weights MBIR algorithm (SAW-MBIR) that is designed to achieve the benefits of both half and full-scan reconstructions in order to maximize temporal resolution over the heart region while providing stable results over the whole volume covered with the wide-area detector. Spatially-adaptive sinogram weights applied to each projection measurement in SAW-MBIR are designed to selectively perform backprojection from the full and half-scan portion of the sinogram based on both projection angle and reconstructed voxel location. We demonstrate with experimental results of SAW-MBIR applied to whole-heart cardiac CT clinical data that overall temporal resolution matches half-scan while full volume image quality is on par with full-scan MBIR.

Motivation & Objective

• To address the conflict between high temporal resolution (requiring half-scan data) and stable image quality (requiring full-scan data) in wide-cone cardiac CT.
• To improve image quality in non-heart regions where half-scan data leads to missing frequencies and artifacts.
• To maintain temporal resolution performance comparable to half-scan reconstruction while reducing artifacts through adaptive data weighting.
• To develop a stable, convergent MBIR framework that handles inconsistent forward/backward projection pairs via spatially-adaptive weights.

Proposed method

• The SAW-MBIR algorithm applies spatially-adaptive sinogram weights that vary based on both projection angle and reconstructed voxel location.
• It performs selective back projection: using half-scan data for voxels in the heart region and full-scan data for voxels outside the heart.
• The method uses a line search to ensure convergence despite inconsistent forward/backward projection operators.
• A cost function is minimized iteratively, with weights dynamically adjusted per voxel to balance data contributions.
• The algorithm is formulated to maintain consistency in the optimization path, ensuring stable convergence to a fixed point.
• The approach integrates statistical modeling and iterative reconstruction to handle incomplete and inconsistent wide-cone projections.

Experimental results

Research questions

• RQ1 Can spatially-adaptive sinogram weights in MBIR improve image quality in wide-cone cardiac CT without sacrificing temporal resolution?
• RQ2 How can full-scan data be leveraged to stabilize reconstructions in non-heart regions while preserving half-scan performance in the heart?
• RQ3 Does the proposed SAW-MBIR method achieve convergence stability despite using mismatched forward/backward projection operators?
• RQ4 To what extent does SAW-MBIR match the image quality of full-scan MBIR and the temporal resolution of half-scan MBIR in clinical whole-heart data?

Key findings

• SAW-MBIR achieves image quality in non-heart regions that matches full-scan MBIR, eliminating the distortions seen in half-scan MBIR.
• In the heart region, SAW-MBIR maintains temporal resolution performance comparable to half-scan MBIR, as confirmed by qualitative and quantitative comparisons.
• Root mean squared error (RMSE) analysis shows SAW-MBIR closely matches full-scan MBIR in non-heart regions and half-scan MBIR in the heart region.
• The method successfully handles inconsistent projections in wide-cone geometry by selectively weighting data based on spatial location.
• The use of a line search ensures convergence stability despite the use of non-uniform, spatially-adaptive weights.
• Clinical results on GE Revolution CT data confirm that SAW-MBIR enables high-quality whole-heart reconstruction with optimal trade-offs between temporal and spatial resolution.