Inter-vendor harmonization of Computed Tomography (CT) reconstruction kernels using unpaired image translation

TL;DR

This paper proposes an unpaired image translation method using a multipath cycle GAN to harmonize CT reconstruction kernels across vendors, reducing measurement differences in quantitative analysis without requiring paired scans.

Contribution

It introduces a novel unpaired harmonization approach with a multipath cycle GAN for cross-vendor CT kernel conversion, eliminating the need for paired training data.

Findings

Harmonization reduces differences in emphysema measurements.

The method highlights the impact of demographic and vendor factors on quantification.

Effective cross-vendor kernel harmonization demonstrated on lung CT scans.

Abstract

The reconstruction kernel in computed tomography (CT) generation determines the texture of the image. Consistency in reconstruction kernels is important as the underlying CT texture can impact measurements during quantitative image analysis. Harmonization (i.e., kernel conversion) minimizes differences in measurements due to inconsistent reconstruction kernels. Existing methods investigate harmonization of CT scans in single or multiple manufacturers. However, these methods require paired scans of hard and soft reconstruction kernels that are spatially and anatomically aligned. Additionally, a large number of models need to be trained across different kernel pairs within manufacturers. In this study, we adopt an unpaired image translation approach to investigate harmonization between and across reconstruction kernels from different manufacturers by constructing a multipath cycle generative adversarial network (GAN). We use hard and soft reconstruction kernels from the Siemens and GE vendors from the National Lung Screening Trial dataset. We use 50 scans from each reconstruction kernel and train a multipath cycle GAN. To evaluate the effect of harmonization on the reconstruction kernels, we harmonize 50 scans each from Siemens hard kernel, GE soft kernel and GE hard kernel to a reference Siemens soft kernel (B30f) and evaluate percent emphysema. We fit a linear model by considering the age, smoking status, sex and vendor and perform an analysis of variance (ANOVA) on the emphysema scores. Our approach minimizes differences in emphysema measurement and highlights the impact of age, sex, smoking status and vendor on emphysema quantification.