Coming theses from other universities

Background: During magnetic resonance imaging (MRI), the interaction between the alternating currents of the gradient coil and the static magnetic field (SMF) generates vibrational forces, perceived as loud acoustic noise. The sound pressure levels (SPLs) are high enough to cause hearing damage, necessitating hearing protection for anyone in the scanner room. To address this, technological advancements have introduced acoustic noise reduction (ANR) software that can alter the gradient currents, resulting in lower vibrational forces and quieter scans. However, such alterations might decrease image quality and prolong scan times. Therefore, it is crucial to evaluate different ANR software to understand their specific utilities and limitations. Additionally, it remains unclear whether, when, and how such software is being utilized clinically, and how MR personnel manage acoustic noise overall. Furthermore, exposure to the SMF within the scanner vicinity can induce transient adverse health effects, such as vertigo, dizziness, nausea, headaches, drowsiness, and metallic taste. However, the evidence is inconclusive regarding the regularity of symptoms, and control groups are needed to account for environmental confounders. Insights into the management of acoustic noise, gradient field applications, and SMF exposure may assist in making MRI both safer and quieter, thereby improving the MR work environment, patient comfort, and the overall clinical experience.

Aim: The overall aim of this thesis was to present a radiography perspective on acoustic noise management, gradient field applications and SMF exposure in MR.

Methods: This thesis encompasses four studies. In Study I, we explored health complaints subjectively associated with SMF and acoustic noise exposure, including symptom prevalence and attribution over the last year. Data were extracted from a nationally distributed cross-sectional survey answered by MR radiographers and CT radiographers (the latter used as controls; CT, computed tomography) across Sweden. In total, data from 529 participants were included and analyzed both descriptively and analytically using logistic regression. In Study II, we conducted 15 individual semi-structured interviews with MR radiographers across Sweden, to explore how they manage acoustic noise in clinical MR settings. The interviews were analyzed thematically. Studies III and IV were both experimental studies that compared ANR software to conventional (non-ANR) imaging. Study III evaluated T2-weighted turbo spin echo (T2W TSE) during lumbar MRI at 1.5 Tesla (T) using two different ANR software – Whisper Mode (WM) and Quiet Suite (QS). In Study IV, we evaluated T2W fast spin echo (FSE) and three-dimensional T1-weighted turbo field echo (3D T1W TFE) during brain MRI at 7 T using the ANR software SofTone. In both Studies III and IV, peak SPLs, perceived noise levels, image quality, and inter-observer agreements between radiologists were compared. In Study III we also measured switched gradient field exposure and gradient currents. Study III included 40 patients, and Study IV 28 healthy volunteers. Data were analyzed using repeated measures analysis of variance (ANOVA), Friedman’s ANOVA, and Wilcoxon’s signed-rank test in Study III. In Study IV, we used paired t-test and Wilcoxon’s signed-rank test. Inter-observer agreement between radiologists’ assessments of image quality was reported in percentage agreement for both studies, with Krippendorff’s alpha also calculated in Study IV.

Results: In Study I, no significant differences in symptom prevalence were seen between the radiographers who work in MR and those in CT. However, working at ≥3 T doubled the risk of SMF-associated symptoms as compared to working at ≤1.5 T. Stress was a significant confounder of symptoms. Work-related acoustic noise was rated as more troublesome by CT than MR radiographers. Study II yielded three main themes on how MR radiographers manage acoustic noise: (I) Navigating Occupational Noise: Risk Management and Adaptation; (II) Protecting the Patient and Serving the Exam, and (III) Establishing a Safe Healthcare Environment with Organizational Support. Compared to conventional imaging, software-based ANR reduced the peak SPLs by 72% and 84% with WM and QS, respectively (Study III), and by 89% (T2W FSE) and 92% (3D T1W TFE) using SofTone (Study IV). In both Studies III and IV, all ANR sequences were perceived as being significantly quieter than their conventional counterparts. Furthermore, all sequences remained diagnostic, although qualitative assessment scores differed slightly for 3D T1W TFE. Switched gradient field exposure was reduced by 48% and 66% using WM and QS, respectively. 

Conclusion: No significant differences in symptom prevalence were seen between MR and CT radiographers. However, working at 3 T increased the risk of SMF symptoms, and stress increased adverse health effects. Although noise was considered to be more troublesome by CT radiographers, this does not suggest that acoustic noise is not problematic in MR environments. MR radiographers demonstrate a willingness to adapt and employ measures to mitigate acoustic noise, thereby enhancing safety and comfort for both patients and personnel, all while striving to deliver diagnostic exams. However, they require appropriate tools and support to do so, suggesting that organizations need to adopt more proactive, holistic approaches to safety initiatives. This thesis also evaluated software-based ANR for its functionality in clinical practice, and found that T2W TSE/FSE and 3D T1W TFE can maintain acceptable subjective image quality while considerably decreasing peak SPLs. This reduction in sound intensity provides both a safer, quieter, and (presumably) more comfortable scan environment.