Dual energy CT with one full scan and a second sparse-view scan using structure preserving iterative reconstruction...

Area of Science:

• Medical Imaging
• Computed Tomography
• Image Reconstruction

Background:

• Conventional dual-energy CT (DECT) requires two full projection datasets, increasing radiation dose and cost.
• Developing efficient DECT reconstruction methods is crucial for reducing patient exposure and improving accessibility.

Purpose of the Study:

• To propose and evaluate a novel iterative algorithm for DECT reconstruction using a reduced data acquisition scheme.
• To enable DECT imaging with one full scan and one sparse-view scan, aiming to lower radiation dose and engineering costs.

Main Methods:

• Developed Structure Preserving Iterative Reconstruction (SPIR), an iterative algorithm utilizing a similarity matrix derived from a full-view scan.
• The algorithm reconstructs a second CT image from sparse-view projections by minimizing image total variation under a data fidelity constraint.
• Evaluated SPIR on digital and physical phantoms, comparing it against Filtered-Backprojection (FBP), Total-Variation-Regularization (TVR), and Prior-Image-Constrained-Compressed-Sensing (PICCS).

Main Results:

• SPIR with a 10-view sparse scan achieved a one-order-of-magnitude reduction in image noise standard deviation (STD) with comparable spatial resolution to full-view FBP.
• SPIR significantly improved reconstruction accuracy over TVR for a 10-view scan (1.33% vs. 6.18% error).
• SPIR outperformed TVR and PICCS in spatial resolution and noise reduction, achieving 7 times lower noise STD than 20-view PICCS with similar resolution.

Conclusions:

• SPIR enables high-quality DECT reconstruction from a combination of full and sparse-view scans.
• The method offers substantial improvements in noise reduction and reconstruction accuracy, potentially reducing imaging dose and cost.
• SPIR demonstrates significant advantages over existing methods, paving the way for more efficient DECT applications.