ESTRO 2024 - Abstract Book

S3896

Physics - Image acquisition and processing

ESTRO 2024

Purpose/Objective:

Dynamic contrast-enhanced (DCE) MRI is a particularly promising imaging biomarker as it can highlight early functional changes in the tumor microenvironment with potential to adapt treatment for personalized biological response. But critical challenges from lack of tools to accurately quantify the MRI images, and lack of consistency across radiotherapy clinics impedes the use of MRI biomarker targeting. The Quantitative Imaging Biomarkers Alliance (QIBA) recommendations for DCE are limited to consistency testing of T1 [2], although they note several attempts at active flow phantoms with pumps or a perfusion phantom with membrane to measure dynamics [3, 4, 5]. The objective of this study is to develop a dynamic phantom to provide reproducible quantification of TOFTs modelling of the permeability kinetics of dynamic contrast enhancement imaging [1].

Material/Methods:

In this study we present a gel based phantom capable of simulating customizable wash in and wash out curves using a programmable motion phantom which is suitable for testing the quantitative generation of permeability kinetics k trans , k ep and v e across any platform of MRI. Feasibility testing of this system was performed on a 1.5T Phillips Ingenia scanner. Layered Gadolinium (Gd) gelatine based gel at different concentrations, to provide different image intensities and T1 weightings, was loaded into the motion insert of the Computerised Imaging Reference Systems (CIRS, Norfolk, USA) “Zeus“ MR-CT motorised abdomen phantom. The layered gel was moved in the superior-inferior direction to simulate a wash in/wash-out curve in the axial plane, as shown in Figure 1. The concentration of Gd was chosen to match the signal intensity in patient data, and the dynamics of the wash in/wash-out curve was also chosen to match dynamics from patient data. The DCE motion phantom was used to compare the permeability kinetics of two options; a high spatial resolution 3D T1 DCE sequence sampling every 21 seconds over 7 minutes, and the high spatial resolution DCE sequence combined with an ultrafast low spatial resolution dynamic contrast sequence sampling every 4 seconds over the first minute. To combine the sequences, the higher spatial resolution scan was resampled to match the ultrafast sequence, and intensity scaling of the sequence was required. The TOFTs analysis was performed on both options using the Phillips Intellispace software permeability package. Model based arterial input function (AIF) was used with the hospital settings for a 1.5T Intravenous injection of gadolinium chelates.

Figure 1 Simulating DCE enhancement kinetics of contrast agent uptake and washout in a gel-based target in an abdomen phantom. (a) shows the motion driver and phantom, (b) shows the movement of the layered gel insert to create contrast wash in/out