Ultrahigh-field MRI is pushing the boundaries of imaging science and has the scope to improve clinical care in specific, particularly neurological, applications. However, the consensus view is there is still some way to go before the radiology community more widely adopts systems with a field strength of 7 Tesla and higher.

“There is high potential with ultrahigh-field MRI, but due to various technical challenges it has not yet reached a high level of confidence within the radiological community,” remarked Prof. Andrew Webb, director of the Gorter Center for High Field MRI, Leiden University Medical Center (LUMC), the Netherlands.

Prof. Andrew Webb, director of the Gorter Center for High Field MRI, Leiden University Medical Center (LUMC), the Netherlands, will speak about challenges and solutions in ultrahigh-field MRI.

Ultrahigh-field corresponds to an MRI system with a magnetic field strength of 7T or above, and there are now thought to be over 100 systems in use worldwide. As the signal-to-noise-ratio linearly increases with the magnetic field strength, 7T MRI offers improved image quality in terms of spatial resolution and contrast. Mainly used as a research tool in the field of neuroinflammatory diseases, 7T MRI has led to the discovery of imaging biomarkers. In a clinical situation with unclear diagnosis but suspected neuroinflammatory disease, ultrahigh-field MRI may be useful to support diagnosis.

From a scientific point-of-view, obtaining high quality images that take full advantage of the theoretical increases in signal-to-noise, spatial resolution, and image contrast is quite a challenge.

“We not only need to overcome intrinsic issues of lower image homogeneity and more localised power deposition in the patient, but have to be very aware that radiologists view images very differently from scientists, particularly with respect to image artefacts,” Webb added.

Use of ultrahigh-field MRI is starting to increase, but it is slow-moving right now. Currently 7T MRI is used clinically in over a dozen hospitals worldwide, including the LUMC, and there will always be a strong investigative component, he continued. Ultrahigh-field is still not a push-button modality and needs highly-trained personnel to run the scans as opposed to lower fields where operation is more routine, and he envisages that ultrahigh-field MRI will be used mainly as an adjunct to 3T in cases where the diagnosis is not clear.

Webb highlighted that the main technical challenges arise from the fact that the images are intrinsically less uniform in terms of signal and contrast across the entire field-of-view of the image, and power deposition can potentially be localised more than at lower field strengths. Also, image artefacts due to motion are much higher.

Brain images acquired at 7 Tesla using a susceptibility-weighted sequence, before and after motion correction. The images on the left show distinct artefacts arising from motion during the 10-minute scan. The images on the right show the improvement possible by incorporating extra information acquired during the scan into the image reconstruction (provided by Prof. Andrew Webb).

“We can help to mitigate these problems using specialised pulse sequences or dielectric materials, by developing accurate models of power deposition, and by devising new methods for correcting images for patient motion, respectively,” he said.

He explained that the main areas of current interest include neurodegenerative and neuromuscular diseases, epilepsy, ocular oncology, and designing acquisition strategies for body imaging with high sensitivity and a high degree of safety. The regulatory landscape is still changing. For instance, the FDA (U.S. Food and Drug Administration) and CE (Conformité Européenne) mark approvals of the Siemens 7 Tesla Terra system in 2017 give scientists and radiologists an opportunity to take more advantage of this technology, he said.

According to Dr. Tim Sinnecker, from the Department of Neurology, Medical Image Analysis Center at University Hospital Basel, Switzerland, the increased spatial resolution and susceptibility effects of ultrahigh-field MRI at 7T have improved imaging of the central nervous system (CNS), where pathophysiological processes take place on a submillimeter scale. For example, the technology can visualise neuroinflammatory lesions in great detail, which has led to the discovery of novel diagnostic and prognostic biomarkers such as the so-called central vein sign.

He added that other imaging findings such as the ‘ring sign’ have direct histopathological correlates (CD68-positive iron-laden microglia or macrophages) and as such have the potential to improve understanding of pathophysiological processes.

In this patient with focal cortical dysplasia in epilepsy, the image on the left was acquired at 3T, and the one on the right at 7 Tesla. The 3T image was judged to be ambiguous in terms of location of the dysplasia, and so the patient was rescanned at 7T. Due to the higher signal-to-noise at 7T, the classic shape of the dysplasia tapering towards the ventricle is clearly visualised (provided by Prof. Andrew Webb).

“The high signal-to-noise ratio renders 7T MRI as an ideal platform for quantitative MRI techniques, which reveal subtle CNS damage not seen on conventional MRI,” he explained.

In addition, Sinnecker noted that 7T MRI is sensitive to iron deposits, myelin, and structural changes of the CNS due to increased magnetic susceptibility effects. Together, these features make 7T MRI an ideal tool to discover novel diagnostic or prognostic biomarkers in neuroinflammatory diseases that will improve future clinical decision-making, he said.

He provided an example of how 7T MRI has been used to identify a biomarker in multiple sclerosis (MS). There is no single specific diagnostic biomarker for this disease.

“It has been known for a century that MS lesions develop around a small central vein. Ultrahigh-field MRI nowadays allows for the detection of this small vein in the centre of MS lesions – the central vein sign. It is both specific and sensitive to MS and may help to differentiate MS from MS imaging mimics such as neuromyelitis optica, Susac syndrome, cerebral vasculitis, or white matter lesions related to migraine or small vessel disease.”

Driven by observations at 7T, researchers are seeking to optimise clinical 3T MRI protocols to make use of the central vein sign in clinical routine. Recent research has confirmed the high specificity of this sign as an MRI biomarker for MS at 3T with clinical protocols, but ultrahigh-field MRI seems to have a higher sensitivity in detecting the sign, asserted Sinnecker.

Joint Session of the ESR and the ESMRMB
Ultrahigh-field (UHF) MRI goes clinical and beyond

  • Challenges and solutions
    Andrew Webb; Leiden/NL
  • The clinical use today
    Tim Sinnecker; Basle/CH
  • New horizons
    Jannie Wijnen; Utrecht/NL


Kolb A, Robinson S, Stelzeneder D et al (2018) Vessel architecture in human knee cartilage in children: an in vivo susceptibility-weighted imaging study at 7 T. Eur Radiol. 28(8):3384-3392: european-radiology.org/5290

Sinnecker T, Oberwahrenbrock T, Metz I et al (2015) Optic radiation damage in multiple sclerosis is associated with visual dysfunction and retinal thinning – an ultrahigh-field MR pilot study. Eur Radiol. 25(1):122-31: european-radiology.org/3358

Liu X, Zhang Z, Zhu C et al (2019) Wall enhancement of intracranial saccular and fusiform aneurysms may differ in intensity and extension: a pilot study using 7-T high-resolution black-blood MRI. Eur Radiol. doi: 10.1007/s00330-019-06275-9: european-radiology.org/6275

Aringhieri G, Zampa V, Marletta M, Biagi L, Tosetti M, Caramella D (2017) Ultra-high-field (7Tesla) MRI study of the articular cartilage in normal subjects. ECR 2017 / C-0244: myESR.org/17244