A Machine-learning Based Initialization for Joint Statistical Iterative Dual-energy CT with Application to Proton Therapy

TL;DR

This paper introduces a CNN-based initialization method to significantly reduce the computational time of iterative dual-energy CT algorithms, improving image accuracy and efficiency for proton therapy applications.

Contribution

A novel CNN-based initialization approach is proposed to accelerate iterative DECT algorithms, demonstrated with DEAM, enhancing accuracy and reducing computation time.

Findings

Improved estimation accuracy for tissue types like adipose, tonsils, and muscle.

Reduced computational time by approximately 5-fold.

Enhanced denoising and image quality in simulated and real data.

Abstract

Dual-energy CT (DECT) has been widely investigated to generate more informative and more accurate images in the past decades. For example, Dual-Energy Alternating Minimization (DEAM) algorithm achieves sub-percentage uncertainty in estimating proton stopping-power mappings from experimental 3-mm collimated phantom data. However, elapsed time of iterative DECT algorithms is not clinically acceptable, due to their low convergence rate and the tremendous geometry of modern helical CT scanners. A CNN-based initialization method is introduced to reduce the computational time of iterative DECT algorithms. DEAM is used as an example of iterative DECT algorithms in this work. The simulation results show that our method generates denoised images with greatly improved estimation accuracy for adipose, tonsils, and muscle tissue. Also, it reduces elapsed time by approximately 5-fold for DEAM to reach the same objective function value for both simulated and real data.