Development and association of new metrics of dose and image quality for comparing and optimizing protocols in CT imaging
Résumé
Introduction. Nowadays, though Computed Tomography (CT) examinations only correspond to a small portion of medical imaging procedures (typically 10 % in France in 2012), they are credited with about 70 % of the total imaging collective dose. Reducing the dose due to CT examinations is therefore a major issue worldwide but cannot be achieved at the expense of the image quality (IQ) needed to ensure a correct diagnosis.
Therefore, CT imaging needs to be described simultaneously using reliable metrics for delivered dose and IQ. However, such metrics are still lacking, especially for IQ. In this study, we developed and combined new metrics of IQ and patient dose in order to compare CT protocols.
Methods. For IQ evaluation, the mathematical model observer (MO) Non Pre-Whitening Eyefilter (Burgess, 1994) was implemented to estimate a detectability index linked to clinical tasks such as lesion detection or discrimination. The index was calculated on CT images of a home-made dedicated phantom acquired for various irradiation and reconstruction conditions. It was then validated in the framework of a clinical study involving a dozen of experienced radiologists, on the large set on images presented in two Alternative Forced Choice (2-AFC) experiments for both tasks. After rescaling on their responses, the MO provided a Percentage of Correct answers (PC) depending on the acquisition parameters, the task and the insert size.
For dose estimation, a complete Monte Carlo model of the GE Discovery CT750 HD scanner was developed with the PENELOPE code. All the components of the scanner were estimated by physical measurements. The modelling was validated with measurements in CTDI and anthropomorphic phantoms using ion chambers and Optically Stimulated Luminescence dosimeters. Once validated, dose levels were calculated in the home-made torso-shaped phantom for every acquisition parameter used in the clinical study. Curves linking PC values calculated by the MO with the simulated dose in the phantom were then built for both detection and discrimination tasks. They thus enabled to represent the influence of irradiation or reconstruction parameters, such as voltage or reconstruction slice thickness.
Results. Values from several CT scanner standard abdominal protocols were placed on the curves after logarithmic regression and compared from the double point of view of dose and IQ. We could then deduce the better protocols in terms of reduced dose while keeping a close PC value. We showed for instance that, according to the clinical task, the patient dose could be reduced by a factor of two keeping a similar probability of correct answers by using a dual-energy protocol with adapted parameters.
Conclusions. This method paves the way for a standardized methodology enabling clinical physicists and radiologists to optimize protocols for defined clinical tasks while keeping the dose as low as possible.
Mots clés
ionizing radiation
metrology
medical physics
Computed Tomography
dosimetry
Medical imaging
CT imaging
Tomography
total imaging collective dose
diagnosis
image quality
delivered dose
IQ
metrics
patient dose
CT protocol
modelling
simulation
IQ evaluation
Non Pre-Whitening Eyefilter
detectability index
signal processing
lesion detection
discrimination
phantom
irradiation
reconstruction conditions
clinical study
dose estimation
Monte Carlo
particle transport
PENELOPE
anthropomorphic phantom
CTDI
ionization chamber
Optically Stimulated Luminescence dosimeter
OSL
dosimeter
dose level
standardization
Domaines
Physique Médicale [physics.med-ph]
Fichier principal
Presentation_ACSimon_2019-06-06 - SFPM - Optimisation protocoles_pour_diffusion_CEA_DOSEO.pdf (1.23 Mo)
Télécharger le fichier
resume_ACSimon_1-s2.0-S1120179719303436-main.pdf (570.95 Ko)
Télécharger le fichier
Origine | Fichiers produits par l'(les) auteur(s) |
---|