Meshless reconstruction technique for digital tomosynthesis.

Area of Science:

• Medical Imaging
• Image Reconstruction
• Digital Tomosynthesis

Background:

• Traditional digital tomosynthesis (DT) reconstruction algorithms often require significant computational resources and memory.
• Existing methods may lack flexibility in slice selection during the reconstruction process.

Purpose of the Study:

• To introduce and evaluate a novel meshless reconstruction algorithm for digital tomosynthesis (DT).
• To assess the algorithm's performance against experimental data and compare it with existing methods.

Main Methods:

• Developed a meshless reconstruction algorithm based on filtered backprojection, adapted for DT geometry.
• Replaced traditional ray tracing with an image mapping procedure for backprojection.
• Implemented two versions: one using convolution filtering and another using Fourier domain filtering.

Main Results:

• The meshless algorithm reconstructs images slice-by-slice, allowing runtime slice position selection.
• The algorithm is memory-efficient and suitable for large datasets.
• Performance was assessed against experimental data and compared favorably with the cone beam algorithm.

Conclusions:

• The novel meshless algorithm provides an efficient and flexible approach to digital tomosynthesis reconstruction.
• Its meshless nature and simplified filtered backprojection adaptation offer advantages in computational demand and usability.
• The algorithm demonstrates comparable or superior performance to existing methods, particularly the cone beam algorithm.