ClinQC: quality control of an X-ray imaging system using clinical images

The work presented in this thesis is part of a research project of Leiden University Medical Center (LUMC) in The Netherlands. It belongs to the field of Diagnostic Radiology analysed from a Medical Physics point of view. After a short overview of the weekly quality controls of an X-ray imaging device, performed using simple phantoms, the thesis focuses on a novel approach called ClinQC (Clinical images-based Quality Control): it has the purpose to monitor the stability of imaging devices, aiming at the early detection of changes in image quality or radiation dose, by deriving quality parameters from chest images of routine patient examinations. The ClinQC algorithm extracts the noise from clinical images and derives the main dose quantities. The noise study presented in this thesis comprehends a validation of the algorithm, performed in several ways: image deteriorations, simulations, phantom studies and real clinical examples. For dose and homogeneity studies only some preliminary results are presented. The thesis collects also some ideas of improvement that can be considered for the future versions of the algorithm and to extend the ClinQC project to other X-ray anatomies and imaging modalities. The obtained similar results for the two compared methods prove that ClinQC is able to give immediate feedbacks of the quality of the imaging devices using patient images. It provides reliable, on-the-fly and sensitive parameters of the quality of the X-ray imaging system, that have the same physical meaning and similar relative variation as the quality indicators of the gold standard QClight method. It can be concluded that the ClinQC algorithm could be already applied in clinical practice, with the initial support of the QClight weekly quality control. In this way, a comparison between the two methods in a real test period will be a guide to find the necessary adjustments of the algorithm until the final version is being installed and stably used in clinical practice.