|Year : 2017 | Volume
| Issue : 2 | Page : 97-102
Brain and spine imaging artefacts on low-field magnetic resonance imaging: Spectrum of findings in a Nigerian Tertiary Hospital
Godwin Ogbole1, Joseph Odo2, Richard Efidi2, Richard Olatunji1, Ayotunde Ogunseyinde1
1 Department of Radiology, University College Hospital; Department of Radiology, University of Ibadan, Ibadan, Nigeria
2 Department of Radiology, University College Hospital, Ibadan, Nigeria
|Date of Web Publication||24-Jul-2017|
Department of Radiology, University of Ibadan, Ibadan
Source of Support: None, Conflict of Interest: None
Background: Low-field (LF) magnetic resonance imaging (MRI) is a technology that is widely used in resource-limited settings for clinical imaging. The images produced, even though of low resolution with noise and artefacts, provide valuable information and guidance for patient assessment and treatment. This study shows a spectrum of MRI artefacts that affect image quality during routine clinical neuroradiology practice using LF MRI in a Nigerian hospital and suggests ways to avoid them. Materials and Methods: We retrospectively reviewed brain and spine MRI studies performed on a 0.36T MagSense 360 (Mindray, China) open MRI at our hospital over a 2-year period to identify image artefacts. About 90% of MRI studies performed at our facility during the study period were neuroimaging. The pattern and distribution of artefacts that featured during imaging were described and illustrative cases demonstrated highlighting their causes and ways to avoid or limit them. Results: Of 936 brain and spine cases evaluated, 506 (54.1%) had artefacts with 369 (72.9%) seen in the brain. Truncation/Gibbs (37.6%) and motion (20.6%) were the most common artefacts in the series, seen most commonly in T2-weighted images. There was no significant difference in the proportion of artefacts between adults and children (P = 0.736). Conclusion: Artefacts are relatively common in neuroimaging with LF MRI and may potentially degrade image quality and interfere with accurate radiological reporting and diagnosis. Improving the recognition of LF MRI artefacts may assist imaging practitioners to avoid or limit their effect on image quality and interpretation.
Keywords: Artefacts, brain, image interpretation, magnetic resonance imaging, spine
|How to cite this article:|
Ogbole G, Odo J, Efidi R, Olatunji R, Ogunseyinde A. Brain and spine imaging artefacts on low-field magnetic resonance imaging: Spectrum of findings in a Nigerian Tertiary Hospital. Niger Postgrad Med J 2017;24:97-102
|How to cite this URL:|
Ogbole G, Odo J, Efidi R, Olatunji R, Ogunseyinde A. Brain and spine imaging artefacts on low-field magnetic resonance imaging: Spectrum of findings in a Nigerian Tertiary Hospital. Niger Postgrad Med J [serial online] 2017 [cited 2020 Mar 29];24:97-102. Available from: http://www.npmj.org/text.asp?2017/24/2/97/211457
| Introduction|| |
Magnetic resonance imaging (MRI) systems are generally few in low-income settings. In recent years, however, they are becoming increasingly more available. Due to lower cost and minimal infrastructural requirement, the low-field (LF) strength systems dominate the market and are majorly used for brain and spine imaging. Understanding the potential artefacts from these systems and their possible clinical implications is essential for improving radiologist interpretations.
An imaging artefact is any structure not normally present but becomes visible as a result of malfunction in the hardware or software of the imaging equipment or a consequence of environmental influences as heat or humidity or generated within a patient or from interactions between the hardware and the patient.
The design of LF scanners makes them susceptible to certain artefacts. These artefacts, however, are not peculiar to LF systems alone. All MRI images may have some degree artefacts in them. A large number are irreversible and may only be reduced, while others can be totally eliminated using appropriate methods and technology. A good knowledge of artefacts is essential for technologists and imaging specialists to maintain optimum image quality. Being adequately informed of the possible artefact, spectrum is extremely vital to avoid or eliminate them and ultimately prevent misinterpretations or misdiagnoses. The continued relevance of LF MRI machines in Sub-Saharan Africa and other low-resource economies with improved technology has led to an overall improvement in the number of modern clinical imaging performed in the region. However, the long acquisition time characteristic of LF MRI machines remains a common cause of artefacts. Radiologists working with MRI are constantly challenged to read images with associated artefacts. It is therefore imperative for them to be able to recognise and correctly identify common artefacts and their possible causes to limit the degradation of MRI quality, avoid repeats and improve the quality of diagnosis. The main aim of this study was to retrospectively identify the common artefacts encountered in our neuroimaging practice within a developing country tertiary hospital, using an LF MRI system, describe them and offer simple strategies for reducing or eliminating them.
| Materials and Methods|| |
We retrospectively reviewed brain and spine MRI studies performed on a 0.36T MagSense 360 (Mindray, China) open MRI at the University College Hospital, Ibadan, Nigeria, from January 2014 to December 2015. Nine hundred and thirty-six neuroimaging studies were available for review on our Picture Archiving and Communication System. All images were reviewed by two senior radiology residents and a consultant neuroradiologist with over a decade experience. Images were reviewed on a workstation with ClearCanvas (version 184.108.40.206TBI), ClearCanvas Inc. Ontario Canada, a standalone diagnostic workstation for image review and utility.
All imaging sequences acquired in each MRI study were evaluated to identify any image artefact. Multiplanar spin-echo T1-weighted (T1W) and T2-weighted (T2W) sequences are routinely obtained in all MRI studies in our centre while fluid-attenuated inversion recovery (FLAIR) and short TI inversion recovery were acquired respectively for all brain and spine studies. Artefacts on imaging studies were first identified by the residents and later verified by the consultant. Classification of artefacts was based on established patterns., We used the grouping system that classified artefacts into two main groups of machine-related artefacts and patient-related artefacts. The different forms of artefacts were entered into a SpreadSheet and analysed using IBM Statistical Package for Social Sciences (SPSS) software version 24 (SPSS Inc. Chicago, IL, USA.). Descriptive analysis was performed on the collected data.
| Results|| |
The MRI examinations performed over the study period (January 2014–December 2015) were reviewed. A total of 1040 MRI examinations were available for review with neuroimaging constituting 90%(936) of the examinations. However, there were 506 cases with artefacts comprising of 269 (53.2%) males and 237 (46.8%) females. The mean age of the patients was 48.3 ± 19.4 years, ranging from <1 to 100 years. Fifty-six (11.1%) patients were in the paediatric age group (<18 years) with mean age of 7.2 ± 5.4 years. Among patients with neuroimaging, 506 (54.1%) had one form of artefact or the other. Brain studies constituted more than two-thirds (72.9%) of all the neuroimaging studies with artefacts. In the spinal region, artefacts were seen in cervical (7.3%), thoracic (3.6%) and lumbosacral regions (7.5%) [Table 1].
There were a significantly higher proportion of artefacts seen on the brain (72.9%) as compared to spine images (27.1%) (P < 0.001).
Most of the patients (64.4%) had artefacts on only one MRI sequence while 180 (35.6%) patients had artefacts on more than one sequence. Artefacts occurred most commonly on T2W sequences while the lowest number of artefacts was recorded on FLAIR images. Overall, Truncation/Gibbs artefacts were most common and these occurred mainly on T2W sequences. Cerebrospinal fluid (CSF) flow artefacts were the most common artefacts on FLAIR images while motion artefacts were the most common artefacts on T1W images [Table 2].
|Table 2: Spectrum of artefacts detected on a low-field magnetic resonance imaging system|
Click here to view
The spectrum of artefacts encountered is demonstrated in [Figure 1],[Figure 2],[Figure 3],[Figure 4],[Figure 5],[Figure 6],[Figure 7],[Figure 8],[Figure 9],[Figure 10].
|Figure 1: (a and b) T2.weighted sagittal and axial brain images showing motion artefact resulting in repeated lines along the cranial vertex across the cerebral hemispheres|
Click here to view
|Figure 2: Axial T2-weighted image of the brain showing partial volume artefact in the right temporal lobe|
Click here to view
|Figure 3: Axial T2-fluid-attenuated inversion recovery image in a 10-month-old patient with communicating hydrocephalus showing a curvilinear hyperintensity in the fourth ventricle consistent with a cerebrospinal fluid flow artefact in the fourth ventricle|
Click here to view
|Figure 5: Coronal T1 images showing susceptibility artefact in the scalp from hair cream product and ventriculoperitoneal shunt tube|
Click here to view
|Figure 6: T1.weighted coronal. (a) T2.weighted sagittal. (b) images of the brain showing Gibbs artefact appearing as bright or dark lines parallel to the curvature of the cranial cavity|
Click here to view
|Figure 7: T2-weighted sagittal brain image showing prominent non-uniformity of the signal across the cerebral hemisphere consistent with radiofrequency inhomogeneity artefact|
Click here to view
|Figure 8: Ghosting artefact seen on (a) T2 sagittal image around orbital globe (b) T1 sagittal at the cranial vault (c) T2 axial of the carotid vessels|
Click here to view
|Figure 9: T2 axial image showing a zipper artefact running across the axial T1 images in the midline and periphery|
Click here to view
|Figure 10: Axial T2-weighted image of the lumbar spine showing slice overlap artefact|
Click here to view
The cause or appearance of commonly encountered artefacts with suggested ways to resolve or limit them is outlined in [Table 3].
|Table 3: Outline identifying cause and resolution of magnetic resonance imaging artefacts|
Click here to view
| Discussion|| |
The classification of MR systems into high-field (HF), mid-field and LF has been regarded as imprecise. Nevertheless, LF MR means field strengths of <0.5T systems., Our series is a retrospective study of brain and spine imaging artefacts identified with an LF MRI system at a tertiary hospital in Nigeria. Practically, every MRI examination has some kind of artefacts, more so an LF system with its inherent low image resolution for clinical diagnostic usage.,
It is understandable that truncation and motion artefacts were most common in this series. Truncation (Gibbs) artefacts are a result of insufficient data sampling which may occur in frequency and phase encoding direction. To save time in LF MR imaging, a smaller number of phase encoding steps may be used, allowing Gibbs to be more commonly observed.
Motion artefacts occur independently of field strength, but the very fast sequences of HF MR more readily aid motion artefact reduction.
Partial volume artefact, which occurs when tissues of different signal intensity become part of the same voxel, are more frequently seen in LF MR than HF MR systems, as thicker image slices are used for acquisition in LF MR. Nonetheless, susceptibility artefacts are less marked in LF than HF MR as noted in our assessment.
These series demonstrate the spectrum of artefacts commonly encountered in the course of neuroimaging with LF MRI. Many classification systems for artefact types have been proposed by several authors;,,, however, we preferred the system which classified artefacts into two main groups of machine- and software-related artifacts and patient-related artefacts.
The machine- and software-related artefacts include zipper, aliasing/wraparound, magnetic susceptibility, chemical shift and truncation/Gibbs artefacts, while the patient-related artefacts include among others, majorly motion and flow artefacts.
All these important artefacts were demonstrated in our LF MRI system and cut across all commonly employed sequences, except for the non-existent chemical shift artefact in our series. Chemical shift artefact is rarely seen with LF systems. The spatial misregistration of fat signal at tissue borders is directly proportional to the field strength and therefore much more noticeable in HF MR imaging. It occurs due to the difference in the resonating frequencies of fat and water.
We have attempted to enumerate the origin, appearance and ways to limit or avoid the commonly encountered artefacts in LF systems in an easy to follow fashion [Table 3]. Our findings indicate that artefacts may present as major imaging constraint and should be adequately anticipated and measures developed to address them for improved image quality and to reduce the chances for misinterpretation. The recognition of an artefact is critical to handle neuroimages, and taking time to understand the varying complexities of image acquisition is important for achieving proper interpretation, avoiding confusion and misdiagnosis. Artefacts have been known to occur with a myriad of appearances depending on the cause or the region being imaged. Brain or spine imaging artefacts may occur at any stage of image preparation, ranging from acquisition to the post-processing period. Artefacts may also result from hardware problems (calibration, power stability), software glitches (programming errors), physiological phenomena (motion, blood flow) and physical limitations (Gibbs and metal susceptibility), especially in the neck and maxillofacial regions, affecting diagnostic quality or present as a reportable pathologic lesion or process.
In our LF system series, we find artefacts [Figure 1],[Figure 2],[Figure 3],[Figure 4],[Figure 5],[Figure 6],[Figure 7],[Figure 8],[Figure 9],[Figure 10] which are comparable to imaging artefacts from HF systems; however, we found a relatively lower proportion of CSF flow artefacts on T2-FLAIR images as compared to those usually reported in HF systems. Patient factors are an important consideration in the development or aetiology of artefacts. Size and physiological motion or processes require that measures are usually taken to forestall their interference with image quality; nevertheless, age does not seem to be associated with artefacts in our series. The use of sedatives in children may account for the lack of age disparity. Achieving paediatric brain images of diagnostic quality could be problematic because of young age or neurological impairment. We routinely sedate paediatric patients for MRI examinations to avoid motion artefacts. Conscious sedation is an acceptable standard practice in neuroimaging with MRI.,, Newer ultra-fast techniques have shorter imaging acquisition times as compared to standard MRI and have reduced movement artefact compared with standard MRI in unsedated children.
The long acquisition time of LF systems remains a primary contributor to motion artefacts and the necessity for sedation. Similarly, Essig et al. reported that flow artefacts are not associated with patients age. However, ventricular size may contribute to the insufficient CSF nulling often reported. In studies performed on a 3T systems, artefacts on T2-FLAIR images were identified more frequently than in 1.5T. The greater the field strength, the more likely CSF flow artefacts develop, especially if there is ventricular dilatation., This is consistent with the few cases of CSF flow artefacts in our series. CSF flow artefacts could be quite large and extending beyond the ventricular system at 3T to overlay adjacent structures and may be unpredictable as they could appear in more unusual locations compared to observations at 1.5T or lower.
| Conclusion|| |
A good understanding of neuroimaging MRI artefacts is imperative for achieving optimal image quality and improving reporting efficiency among radiologists working with LF MRI in resource-limited settings. We attempted to provide few suggestions for reducing these artefacts; however, more comprehensive technical descriptions of procedures artefact prevention may be found in cited literature.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Potchen MJ, Kampondeni S, Birbeck GL, Hammond CA, Gonani A, Phiri KS, et al.
Magnetic resonance imaging in Malawi: Contributions to clinical care, medical education and biomedical research. Malawi Med J 2011;23:60-4.
Ogbole G, Adeleye A, Adeyinka A, Ogunseyinde O. Magnetic resonance imaging: Clinical experience with an open low-field-strength scanner in a resource challenged African state. J Neurosci Rural Pract 2012;3:137-43.
] [Full text]
Krupa K, Bekiesinska-Figatowska M. Artifacts in magnetic resonance imaging. Pol J Radiol 2015;80:93-106.
Kathiravan S, Kanakaraj J. A review on potential issues and challenges in MR imaging. Scientific World Journal 2013;2013:783715.
Ogbole GI, Adeyinka OA, Okolo CA, Ogun AO, Atalabi OM. Low field MR imaging of sellar and parasellar lesions: Experience in a developing country hospital. Eur J Radiol 2012;81:e139-46.
Somasundaram K, Kalavathi P. Analysis of imaging artifacts in MR brain images. Orient J Comput Sci Technol 2012;5:135-41.
Morelli JN, Runge VM, Ai F, Attenberger U, Vu L, Schmeets SH, et al.
An image-based approach to understanding the physics of MR artifacts. Radiographics 2011;31:849-66.
Hayashi N, Watanabe Y, Masumoto T, Mori H, Aoki S, Ohtomo K, et al.
Utilization of low-field MR scanners. Magn Reson Med Sci 2004;3:27-38.
Martí -Bonmatí L, Kormano M. MR equipment acquisition strategies: Low-field or high-field scanners. Eur Radiol 1997;7 Suppl 5:263-8.
Ishak NF, Logeswaran R, Tan WH. Artifact and noise stripping on low-field brain MRI. Int J Biol Biomed Eng 2008;2:59-68.
Arena L, Morehouse HT, Safir J. MR imaging artifacts that simulate disease: How to recognize and eliminate them. Radiographics 1995;15:1373-94.
Taber KH, Herrick RC, Weathers SW, Kumar AJ, Schomer DF, Hayman LA. Pitfalls and artifacts encountered in clinical MR imaging of the spine. Radiographics 1998;18:1499-521.
Stadler A, Schima W, Ba-Ssalamah A, Kettenbach J, Eisenhuber E. Artifacts in body MR imaging: Their appearance and how to eliminate them. Eur Radiol 2007;17:1242-55.
Bellon EM, Haacke EM, Coleman PE, Sacco DC, Steiger DA, Gangarosa RE. MR artifacts: A review. AJR Am J Roentgenol 1986;147:1271-81.
Lavdas E, Tsougos I, Kogia S, Gratsias G, Svolos P, Roka V, et al.
T2 FLAIR artifacts at 3-T brain magnetic resonance imaging. Clin Imaging 2014;38:85-90.
Adamietz B, Cavallaro A, Radkow T, Alibek S, Holter W, Bautz WA, et al.
Tolerance of magnetic resonance imaging in children and adolescents performed in a 1.5 Tesla MR scanner with an open design. Rofo 2007;179:826-31.
Boswinkel JP, Litman RS. Sedating patients for radiologic studies. Pediatr Ann 2005;34:650-4, 656.
Treluyer JM, Andre C, Carp PF, Chalumeau M, Tonnelier S, Cuq C, et al.
Sedation in children undergoing CT scan or MRI: Effect of time-course and tolerance of rectal chloral hydrate. Fundam Clin Pharmacol 2004;18:347-50.
Salvo I, Cucchi C, Camporesi A, Borghi B, Grasso F, Silvani P. Sedation with magnetic resonance imaging in pediatric patients. Minerva Anestesiol 2002;68:414-9.
Woodfield J, Kealey S. Magnetic resonance imaging acquisition techniques intended to decrease movement artefact in paediatric brain imaging: A systematic review. Pediatr Radiol 2015;45:1271-81.
Essig M, Hawighorst H, Schoenberg SO, Engenhart-Cabillic R, Fuss M, Debus J, et al.
Fast fluid-attenuated inversion-recovery (FLAIR) MRI in the assessment of intraaxial brain tumors. J Magn Reson Imaging 1998;8:789-98.
Ogbole GI, Soneye MA, Okorie CN, Sammet S. Intraventricular cerebrospinal fluid pulsation artifacts on low-field magnetic resonance imaging: Potential pitfall in diagnosis? Niger Med J 2016;57:59-63.
] [Full text]
Bakshi R, Caruthers SD, Janardhan V, Wasay M. Intraventricular CSF pulsation artifact on fast fluid-attenuated inversion-recovery MR images: Analysis of 100 consecutive normal studies. AJNR Am J Neuroradiol 2000;21:503-8.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10]
[Table 1], [Table 2], [Table 3]