Automating routine CT image quality analysis from a virtual server

Abstract

Routine QA is an effective approach to monitoring CT performance over time, it consists of a series of image quality checks which give the physicist the ability to detect suboptimal or declining performance and initiate corrective actions. This contributes to high quality images that improve the likelihood for correct diagnosis and ultimately contributes to quality patient care. With the clear importance of routine CT QA the drawback lies in the tedious and time consuming image analysis measurements, likewise in busy radiology departments it can be difficult schedule time for QA on the CT. In response to these problems we have developed and introduced an effectively automated QA computer program that automatically executes image quality analysis. The QA program platform is hosted on a virtual server and was scripted in-house using a combination of MatLab, PowerShell and ConQuest DICOM Server. With the introduction of this QA program, the routine CT QA procedure has been compressed down to scanning the QA phantom and exporting the study straight from the console via a dedicated Physics DICOM node. The QA program automatically begins image analysis then saves results, archives images and emails the user with the outcome of the analysis. With the introduction of this QA program we have observed a significant reduction in scanner downtime, an increase in the number of images analysed and improved consistency of measurements. The overall result has been very positive with improved analysis of scanner performance and improved efficiency of routine CT QA.