Optimization of acoustic radiation force imaging: Influence of timing parameters on sensitivity

• Published in: Magnetic resonance in medicine Vol. 79; no. 2; pp. 981 - 986
• Main Authors: Dadakova, Tetiana, Krafft, Axel Joachim, Özen, Ali Caglar, Bock, Michael
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.02.2018
• Subjects: Acoustic noise; acoustic radiation force imaging (ARFI); Coding; elastic tissue properties; Elasticity Imaging Techniques - methods; Humans; Image Processing, Computer-Assisted - methods; Magnetic resonance; Magnetic Resonance Imaging - methods; Models, Biological; Motion compensation; motion encoding gradients; MR pulse sequence optimization; MR‐guided focused ultrasound; Optimization; Parameter sensitivity; Phantoms, Imaging; Phase transitions; Radiation; Sensitivity; Sensitivity analysis; Signal-To-Noise Ratio; Sound waves; Ultrasound; acoustic radiation force imaging (ARFI); elastic tissue properties; MR pulse sequence optimization; MR-guided focused ultrasound; motion encoding gradients
• ISSN: 0740-3194; 1522-2594
• DOI: 10.1002/mrm.26734

Purpose Optimization of timing parameters for MR‐guided ARFI to achieve the highest displacement signal‐to‐noise ratio (SNRd). Theory and Methods In MR‐guided ARFI the phase change induced by motion encoding gradients (MEGs) is measured to assess tissue displacement. The sensitivity of this encoding procedure depends on several timing parameters, such as the MEG duration and the offset time between ultrasound (US) and MEG. Furthermore, mechanical and MR tissue constants and MEG schemes (bipolar or three‐lobed) influence SNRd. Optimal timing parameters were determined in simulations for bipolar and three‐lobed MEGs, and the results were compared with measurements. To provide clinically usable timing parameters, physiologically relevant ranges of tissue constants were considered. Results For the considered ranges of tissue constants, optimal timing parameters provide only 6% higher SNRd for bipolar than for three‐lobed MEG. Three‐lobed MEG is less sensitive to motion as confirmed in phantom experiments. Bipolar MEG can use approximately 1.5‐fold shorter MEG durations. Conclusion Both bipolar and three‐lobed MEGs can yield approximately the same SNRd if the optimal timing parameters are chosen. Bipolar MEG allows for shorter durations, which is preferable if deposition of US energy needs to be minimized, and three‐lobed MEG is more suitable when residual motion compensation is necessary. Magn Reson Med 79:981–986, 2018. © 2017 International Society for Magnetic Resonance in Medicine.