Reducing zebra streak artifact in 3D TSE C-spine MRI images by manipulating the position of saturation bands

Abstract Description

Presentation Upload :

Authors (including presenting author) :

Chiu WK(1), Chan WM(1),Yeung L(1)

Affiliation :

(1)X-Ray Department, Our Lady of Maryknoll Hospital

Introduction :

Artifacts such as truncation artifacts, wrap around artifacts and motion artifacts are common drawbacks of MRI which limits the diagnostic values of the images. In a study, artifacts were present in 38% of overall MRI scans (Budrys et al, 2018). Although experienced radiologists are able to distinguish those artifacts from normal anatomy or pathology, radiographer should still attempt to provide the best possible image quality so that all anatomical structures are demonstrated clearly in the limited scanning time. 3D T1-weighted TSE is one of the sequences in MRI c-spine protocol. However,zebra stripe artifact were spotted obscuring part of the cord, vertebral body and facet joints on sagittal images. It maybe caused by saturation pulse interfering with the data acquisition in 3D-TSE scans. Countermeasures include changing the number of saturation bands, decreasing the TSE factor and using an even number for number of signal averages (NSA) (Philips, 2004). However, those measures can potentially increase the time for image acquisition.

Objectives :

In the image planning for 3D T1-weighted TSE, the saturation bands were placed besides the soft tissue neck using a coronal survey image. If zebra stripe artifact is caused by the interference of saturation band during 3D-TSE image acquisition, the objective of this study is to investigate the feasibility of eliminating the artifacts by manipulating the position of the saturation bands without changing other parameters and image acquisition time.

Methodology :

All patients undergo c-spine MRI scans in Kwong Wah Hospital Philips Achieva 1.5T MRI scanner from January to March and from June to August in 2018 were included. From January to March, old setting (saturation bands placed just besides area of interest) was applied for all patients. From June to August, new setting (saturation bands placed 20mm away from the area of interest to cover both shoulders) was applied. Sagittal 3D T1-weighted TSE images were reviewed to observe if zebra stripe artifacts were present.

Result & Outcome :

From January to March, 30 C-spine MRI scans were performed. The mean age of the patient was 58 (max: 94, min:30), including 13 males (43.3%) and 17 females (56.7%). Clinical indications for MRI C-spine included 25 limb numbness (83.3%), 4 neck pain (13.3%) and 1 cord compression (3.4%). Zebra stripe artifacts were spotted in 30 3D-TSE images (100%). From June to August, 43 C-spine MRI scans were performed. The mean age of the patient was 62.9 (max: 88, min:41), including 16 males (37.2%) and 27 females (62.8%). Clinical indications for MRI C-spine included 32 limb numbness (74.5%), 5 neck pain (11.7%) and 2 cord compression (4.6%), 2 C1/C2 subluxation (4.6%) and 2 trauma (4.6%). Zebra stripe artifacts were spotted in 13 3D-TSE images (30%). In this study, 70% of zebra stripe artifact reduction was noted when new setting was utilised. Radiographers have to strike a balance between image quality and image acquisition time to produce the best possible diagnostic images in a time-efficient way. The study revealed that method as simple as manipulating the position of the saturation bands in 3D-TSE can be a good alternative for effectively reducing zebra stripe artifact to improve diagnostic value of MRI images without compromising image acquisition time and image quality in terms of resolution and signal-to-noise ratio. Future investigation can compare the effectiveness of the routine counter measures mentioned above and this alternative setting for eliminating zebra stripe artifact.