# A streamlined workflow for comprehensive 4DCT imaging QA with automated analysis of spatial integrity and image quality

## TL;DR

This paper introduces an automated workflow for 4DCT imaging quality assurance, improving efficiency and consistency in evaluating spatial integrity and image quality.

## Contribution

The novel contribution is a scalable, automated QA workflow using a widely available phantom and TotalQA software for 4DCT imaging.

## Key findings

- Automated analysis reduced QA time by over 80% compared to manual methods.
- Spatial integrity was within ±1 mm and HU values within two HU across all motion phases.
- Slight increases in image noise were observed in 4DCT compared to stationary CT.

## Abstract

A comprehensive 4DCT QA program that includes the assessment of spatial integrity and image quality of 4D phantom scans can be resource‐intensive, especially because the analysis burden scales with the number of motion traces used for QA. This work presents a streamlined and scalable workflow, enabled by the use of a widely available phantom and an automated analysis tool.

For 4DCT imaging QA, the Catphan was placed on the QUASAR motion platform, driven with sinusoidal motion traces of various amplitudes and frequencies. The acquired image sets were automatically analyzed using a newly designed 4DCT analysis module in TotalQA. The metrics of interest for 4DCT QA were analyzed for image sets of all motion bins and compared to the stationary reference image set. Three broad categories of imaging tests for 4DCT QA were defined: spatial integrity, HU constancy, and image quality. The sensitometry plugs in the Catphan were used as motion surrogates for evaluating spatial integrity, including the dimensions of the plugs and the observed motion amplitude in the scans. HU values of the sensitometry plugs in different phases were evaluated for consistency. For image quality, spatial resolution, low contrast resolution, and image noise were evaluated.

Compared to the stationary reference CT, spatial integrity was within ±1 mm and the HU values were within two HU in the 4DCT for all phases. The spatial resolution was consistent while slightly higher noise was observed in the 4DCT images. The automated analysis in TotalQA was completed in approximately 20 min per motion trace, improving efficiency by more than 80% compared to the manual workflow.

A streamlined 4DCT imaging QA workflow with automated analysis offers a robust assessment of imaging quality and motion accuracy in 4DCT scans, enabling an efficient and consistent QA process in large healthcare networks.