Loud noise during MRI scans is the leading cause of patients' anxiety, but the origin of this loud noise, mainly fast- switching fields, is also an essential component to generate images. Current methods that reduce acoustic noise during scans rely almost solely on slowing down the switching field, which have limitations in both flexibility and sound reduction effect, and result in reduced scan efficiency. By taking advantages of the new degrees of freedom provided by Magnetic Resonance Fingerprinting (MRF), we will develop a systematic framework to 1) reduce patients' anxiety by changing the sound emitted from the scanner to music, which allows the use of fast-switching fields while providing pleasing sound; 2) simultaneously provide quantitative information on all tissue properties needed for a clinical scan and 3) maintain the same high efficiency as compared to conventional (noisy) scans. To this end, first, we will develop an acoustic model to characterize the acoustic response of the scanner system. Second, we will generate the framework that evaluates and optimizes the acoustic response, image quality and scan time from any acoustic input. This framework will be used to optimize 3D MRF scans with 1 mm3 isotropic resolution and with whole brain coverage. Finally, we will validate the framework on a patient study to assess patients' anxiety, image quality, scan efficiency and success rate of recruiting patients who previously refused to have an MRI scan. The results generated in these studies could improve the outcomes of any patients who undergo an MR scan: the quantitative scans will provide more definitive information for lesion characterization in the form of quantitative maps, and the pleasing sound will significantly change patients' experience during the MR scans, which could lead to greater compliance and reduced motion issue, and could break down the barriers that keep children and anxious adults from receiving MRIs without sedation.
Loud noise during MRI scans is the leading cause of patients' anxiety, but the origin of this loud noise, mainly fast- switching fields, is also an essential component to generate images. Previous methods rely almost solely on slowing down the switching field to reduce the noise, resulting in reduced scan efficiency. We propose a general framework that could resolve this longstanding conflict by changing the sound of the MRI scan to music while simultaneously providing multiple quantitative tissue properties with high scan efficiency.