Image quality assessments according to the angle of tilt of a flex tilt coil supporting device: An ACR phantom study

Abstract In this study, we assessed how image quality depends on the angle of tilt of a flex tilt coil supporting device during an MRI examination. All measurements were performed with an American College of Radiology (ACR) MRI phantom using a flex tilt coil supporting device. All images were analyzed using an automatic assessment method following the ACR MRI accreditation guidance. Image quality was compared between acquisitions grouped according to the angle of tilt of the coil supporting device: group A (Flat mode), group B (10˚), and group C (18˚). All measured image qualities were within the ACR recommended criteria, regardless of the angle of tilt of the flex tilt coil supporting device. However, statistically significant differences between the three groups were found for slice thickness, position accuracy, image intensity uniformity, and SNR (P < 0.05, ANOVA). The flex tilt coil supporting device can provide sufficient image quality, passing the criteria of the ACR MRI guideline, despite differences in slice thickness, slice position accuracy, image intensity uniformity, and SNR according to the angle of tilt.


| INTRODUCTION
Modern MRI scanners are generally equipped with multichannel transmit-receive coils of the birdcage design. [1][2][3][4] With the development of parallel imaging techniques, these coils provide uniform radiofrequency fields and spatially uniform image quality with a reduced scanning time. [5][6][7][8] However, these modern coils generally have a fixed-geometry coil volume that is difficult to use with patients with kyphosis of the spine, who cannot lie completely flat during the MRI examination. For patients with such conditions, the conventional coil design causes patient discomfort and increases the rate of MRI examination failures. To address these problems, a nonmetallic flex tilt coil supporting device can provide an alternative geometry for the birdcage coil, allowing easier positioning and scanning of such patients. However, a coil supporting device tilts the coil off the isocenter in the anteroposterior direction, and the isocenter is one of the most important factors affecting image quality because the magnetic field degrades and gradient field nonlinearities increase with distance from the isocenter. [9][10][11] Theoretically, imaging at or near the isocenter is desirable to produce high quality images, but it is not always possible during an MRI examination.

2.B | MR equipment and scan parameters
All images were acquired on a clinical 3.0-T MRI scanner (Ingenia CX; Philips Healthcare, the Netherlands) with an 80 mT/m maximum gradient strength and a 200 T/m/s slew rate. Fifteen-channel head coils (Philips Healthcare) were used for image acquisition. Two axial spin echo sequences were used to acquire T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI), according to the standard sequence protocols prescribed by the ACR guideline. 12 The uniformity correction mode CLEAR (scan option to improve the image intensity uniformity under Philips Healthcare) and adaptive radio-frequency (RF) shimming were used for both T1WI and T2WI at a fixed and adequate bandwidth to reduce degradation of image quality and variation in RF nonuniformity. Both spin echo sequences were acquired in the axial plane based on the phantom frame of reference with the following parameters: field of view: 250 × 250 mm; voxel size: 1 × 1 mm; acquisition matrix: 256 × 256; number of excitations: 1; slice thickness: 5 mm; slice gap: 5 mm; number of slices: 11; receiver bandwidth: 218 Hz/pixel. Further description of the parameters is given in Table 1.

2.C | Image analysis
To analyze the quality of the acquired images, an ACR MRI quality control test consisting of eight quantitative tests was performed on seven sets of scans obtained under the same setup for each tilt angle using an open-source Matlab code (R2016b; Mathworks, Natick, MA, USA) available from http://jidisun.wix.com/osaqa-project/. 13 The signal to noise ratio (SNR) measurement was performed using the subtraction method according to the following equation (an image subtraction was performed to produce a noise only image) 14,15 : where S is the mean signal value of the two images and σ is the standard deviation of the subtracted images. S and σ values were

2.D | Statistical analysis
The Kolmogorov-Smirnov test was used to confirm that the eight For all statistical analyses, a two-sided level of P < 0.05 was considered statistically significant.

| RESULTS
The measurements of the eight image parameters are presented in Table 2. Representative images acquired from all three groups are shown in Fig. 2. For geometric accuracy, all measured values were within the ACR criterion (AE2 mm) for the true values. There were no statistically significant differences between groups A, B, and C in any direction (P > 0.05).
The spatial resolution of both slice 1 images of T1WI and T2WI in both directions passed the ACR criterion of 1.0 mm in all three groups. No statistically significant differences in either slice 1 of T1WI or T2WI in either direction were found between the groups (P > 0.05). For slice thickness accuracy, all measured values in all three groups were within the ACR criterion of 5.0 AE 0.7 mm. There were no significant differences in slice 1 of T2WI between groups A, B, and C (P > 0.05). In slice 1 of T1WI, no statistically significant differences were found between groups A and C (P = 0.052) (Fig. 3).  addition, there was no statistically significant difference between groups B and C in slice 11 of T1WI (P = 0.513). However, there were significant differences between groups A, B, and C for slice 1 of T2WI (P < 0.05) (Fig. 3).
For image intensity uniformity, there were no statistically significant differences in either slice 7 of T1WI or slice 7 of T2WI between groups B and C (P = 0.277 and P = 0.111, respectively) ( Fig. 3). In addition, all measured image intensity uniformities were greater than the ACR criterion of 82% for 3.0 T. For SNR, there were no statistically significant differences in slice 7 of T1WI between groups A, B, and C (P > 0.05) or in slice 7 of T2WI between groups B and C (P = 0.123). For percent-signal ghosting, there were no significant differences between groups A, B, and C (P > 0.05). All measured ghosting ratios were less than the ACR criterion of 0.025.
For low-contrast object detectability, the total number of measured spokes in all three groups passed the ACR criterion of greater than 37 spokes for 3.0 T. There were no statistically significant differences between groups A, B, and C for either T1WI or T2WI (P > 0.05). However, the total number of measured spokes tended to decrease as the angle of tilt increased. When a phantom is used for a quantitative and qualitative image quality analysis, it is important to indicate definite and objective criteria. In addition, the QA process should be easy to perform and as simple and convenient as possible. Previous studies used a standard set of image QA procedures using numerous phantoms. [16][17][18] However, manual assessment methods appear to be complicated and inefficient for assessing image quality, tending to be highly dependent on the observer or monitor, and also time consuming. There may also be the problem that contrast evaluation, which is considered to be a crucial image quality category, is not performed. On the other hand, some studies have demonstrated automatic assessment methods to reduce the QA processing time while improving the consistency and objectivity of measured values. 13,19,20 Thus, we also used automatic image quality measurements available through an open-source code, measurements that were relatively easy to perform in the current study.

| DISCUSSION
Our results showed that most values of the categories evaluated were similar, regardless of the angle of tilt of the flex tilt coil. However, some comparisons revealed statistically significant differences in slice thickness, slice position accuracy, and image intensity uniformity. These can be explained by magnetic field inhomogeneity and gradient field nonlinearity caused by moving away from the isocenter of the magnet bore, with these causing image distortion and degradation. A few studies showed that magnetic field inhomogeneity and gradient field nonlinearity increase toward the periphery away from the magnet isocenter, thus leading to nonuniform intensity. 9

| CONCLUSIONS
The flex tilt coil supporting device can provide sufficient image quality passing the criteria of the ACR MR phantom guideline, despite differences in slice thickness, slice position accuracy, image intensity uniformity, and SNR according to the angle of tilt.

CONFLI CT OF INTEREST
No conflict of interest.

ACKNOWLEDG MENTS
We thank Scientific Publication Team, Asan Medical Center for their contributions to proofreading our paper in English.

A U T H O R C O N T R I B U T I O N
Sung Min Kim guarantors of integrity of entire study. Ji Sung Jang and Ho beom Lee were involved in study concepts and data acquisition. Ho beom Lee and Ki Baek Lee were involved in data analysis and statistical analysis. All authors were involved in literature research.