48 Reference flow and role of medical physicist in the validation of 4D flow MRI protocols. (December 2018)

• Record Type: Journal Article
• Authors: Puiseux, T.; Sewonu, A.; Meyrignac, O.; Nicoud, F.; Rousseau, H.; Mendez, S.; Moreno, R.
• Is Part Of: Physica medica. Volume 56(2018)Supplement 1
• Page Start: 27
• Page End: 28
• Publication Date: 2018-12
• Subjects: Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2018.09.061
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Introduction: Magnetic resonance imaging (MRI) applied to 4D hemodynamics conditions assessment is increasingly used in clinical routine to detect and treat cardiovascular diseases[1, 2] . Although applications are obvious and the power of new technologies (deep learning) respond effectively to this demand (Arterys), the quality of the results remains highly dependent on the parameters of the acquisition MRI sequence[3] . The need to limit the examination time leads to then degrade the sequence parameters, which leads to a loss of control over the quality of the speed measurements. So far, no quality assurance program has been able to assess the quality of 4D flow measurements. In this work, we propose a 'reference' flow intended to estimate deviations with the 4D Flow measurements obtained by MRI. Methods: An experimental device composed by an MRI-compatible pump, flow and pressure sensors and a 3D printed phantom, materializes reference flow. Since flow conditions and geometry are well controlled, 4D flow is correctly modeled by computational fluid dynamics (CFD). The measurements obtained by MRI on the experimental model are compared to modeling. The deviations obtained are evaluated on all the points of the flow for velocity components. Results: Difference between the reference flow (CFD) and idealized 4D Flow MRI sequence (Siemens Avanto, 1.5T, voxel 2 × 2 × 2 mm, 35 phases, 45 min of acquisition) is 0.96% for velocity magnitude measurements. Differences with clinical less accurate sequence (voxel >3 mm, short acquisition time, parallel imaging) are very diverse and can exceed 20% if the sequence parameters are not controlled. Conclusions: Clinical use of 4D Flow MRI sequences must benefit from a quality assurance program initiated by the head of radiology department and the manufacturer. The great variability of acquisitions and devices (brands, models, field) requires that a reference system for quality control be defined and that the implication of the medical physicist in the protocol validation process is established. The proposed reference system is a tool to achieve this goal.