A systematic review of acoustic noise sources in Magnetic Resonance Imaging

Acoustic noise in nuclear Magnetic Resonance Imaging (MRI), with sound pressure levels (SPL) exceeding 130 dB, is an ongoing issue for patient and operator health and well-being, as well as for signal quality. Understanding the noise characteristics and generation mechanism is essential for the accurate design of noise control methods to be applied in the design of new devices and imaging sequences. The aim of this work is to provide an overview of the dominant noise sources in a Nuclear Magnetic Resonance (NMR) scanner, to highlight relationships between noise characteristics (SPL, noise spectrum) and MRI design features (static magnetic field, type of applied sequence and sequence parameters, gradient coil features) and to investigate the current state of the art in terms of analytical and numerical modeling of vibration and noise generation. Available information on the research questions was collected through a systematic approach following a procedure based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, and the Scopus database was used to search the relevant literature. A total of 131 publications qualified for the inclusion in the review; the topics covered by the included papers are: SPL measurements on MRI sequences (N = 77); acoustic and/or structural frequency response function (FRF) measurements (N = 26), analytical models for acoustic noise and/or vibrations (N = 25), numerical models for acoustic noise and/or vibrations (N = 26). It was found that acoustic noise depends on the rapid switching of gradient coils during imaging sequences. It determines time-varying Lorentz forces both on the driving currents of the coil itself and on the eddy currents on the metallic parts of the scanner. The forces field generates mechanical vibrations that act as noise sources. Consequently, SPL increases with increasing static magnetic field B0, gradient coil’s driving current and slew rate, and it’s influenced by the excited mechanical and acoustic modes.