HEAD AND NECK

Feasibility of high-resolution pituitary MRI at 7.0 tesla
Alexandra A. J. de Rotte & Anja G. van der Kolk & Dik Rutgers & Pierre M. J. Zelissen & Fredy Visser & Peter R. Luijten & Jeroen Hendrikse
Received: 3 March 2014 /Revised: 22 April 2014/Accepted: 8 May 2014/Published online: 29 May 2014 # European Society of Radiology 2014
Abstract Objectives Since the pituitary gland measures 3-8 mm, imagingwiththehighestpossiblespatialresolutionisimportantfor the detection of even smaller lesions such as those seen in Cushing’s disease. In the current feasibility study, we tested a multi-sequence MRI protocol to visualize the pituitary gland in high resolution at 7.0 Tesla (7.0 T). Methods Ten healthy volunteers were examined with a 7.0 T pituitary gland protocol. The protocol consisted of a T1weighted magnetization-prepared inversion recovery (MPIR) turbo spin-echo (TSE) sequence and a T2-weighted TSE sequence. Additionally, this protocol was tested in five patients with clinical and biochemical suspicion of a microadenoma. Results The dedicated protocol was successful in visualizing normal pituitary anatomy. At 7.0 T compared to 1.5 T, four timesasmanyslicescoveredthepituitaryglandinsagittaland coronal direction. In three patients, a lesion was diagnosed at 7.0 T, and was confirmed by histopathology to be a microadenoma. Conclusion Head-to-head comparisons of 7.0 T with 1.5 T and 3.0 Tare needed with larger samples of patients and with imaging times feasible for clinical settings. However, the current study suggests that high-resolution 7.0 T MRI of the
pituitary gland may provide new perspectives when used as a second-line diagnostic examination in the specific context of Cushing’s disease. Key Points • 7.0 T MRI enables ultra-high-resolution imaging of the pituitary gland. • 7.0 T MRI isappropriate tovisualize normalpituitary gland anatomy. • The pituitary protocol consists of a T1-MPIR-TSE and a T2TSE sequence. • In four patients, a suspected ACTH-producing microadenoma was visualized at 7.0 T. • Histopathologyconfirmed three offour lesions to beACTHproducing microadenomas.
Keywords Pituitarygland .Cushing’sdisease.Pituitary . Magneticresonanceimaging
Abbreviations ACTH Adrenocorticotropic hormone CNR Contrast-to-noise ratio IRB Institutional review board MPIR Magnetization-prepared inversion recovery MRI Magnetic resonance imaging SAR Specific absorption rate SNR Signal-to-noise ratio T Tesla TFE Turbo field echo TSE Turbo spin echo
Introduction
Magnetic Resonance Imaging (MRI), with its superior soft tissue contrast, is the preferred modality for visualization of
A.A.J. deRotte (*):A. G. van der Kolk:D.Rutgers:F. Visser : P. R. Luijten:J. Hendrikse Department of Radiology, University Medical Center Utrecht, Heidelberglaan 100, Postbox 85500, 3508 GA Utrecht, The Netherlands e-mail: a.a.j.derotte@umcutrecht.nl
P. M. J. Zelissen Department of Internal Medicine (Section of Endocrinology), University Medical Center Utrecht, Utrecht, The Netherlands
F. Visser Philips Healthcare, Best, The Netherlands
Eur Radiol (2014) 24:2005–2011 DOI 10.1007/s00330-014-3230-x
the pituitary gland [1, 2]. This imaging technique is able to achieve high contrast-to-noise ratio (CNR), even without the use of contrast agents, for normal anatomy as well as pathological lesions [3–6]. High-resolution imaging of the pituitary gland is necessary for pathologies such as Cushing’s disease due to the presence of small microadenomas measuring less than 6 mm in the majority of cases [7]. In clinical settings, 1.5 Tesla (T) MRI is commonly used for pituitary imaging. However, very small microadenomas often remain undetected with this standard low-field-strength MRI system due to the low spatial resolution that can be gained within reasonable examination time [8]. Furthermore, partialvolumeeffectsofthesurroundingstructuresmayresult in false-positive diagnosis of microadenomas when using low-resolution MR imaging of the pituitary gland. Higher-field-strength MRI systems – for instance, 3.0 Tor 7.0 T – significantly improve the image quality as a result of higher attainable signal-to-noise ratio (SNR) [4, 5, 9–14]. However, there are several negative consequences associated with higher magnetic field strength [11, 12]. Because the pituitary gland is located in close proximity to the sphenoid sinus,susceptibilityeffectsmayposeaproblemduetotheairbrain interface. Furthermore, the relaxivity of gadoliniumbased contrast agents is lower with increased field strength [15]. The combination of prolonged T1-values and the fast uptake and washout of gadolinium in the pituitary gland may affect the image contrast. Notwithstanding these effects, a series of studies has shown that high-field-strength MR imaging at 3.0 T is superior for detection of suspected pituitary gland pathology [4, 5, 8, 12–14]. MR imaging at 7.0 T remains challenging, however, as the issues associated with the higher field strength may be even more pronounced. Additionally, safety test data for metallic implants at this strength are limited, resulting in most metallic implants being considered as a major contraindication for ultrahigh-field-strength MRI. The aim of this study was to investigate whether it is feasible to visualize the pituitary gland with MRI at the ultrahigh field strength of 7.0 T. For protocol evaluation, 10 healthy volunteers were examined at 7.0 T. In addition, clinical 7.0 T MR images of the pituitary gland of five patients with no clear lesion at 1.5 Twere shown.
Materials & methods
Study population
Institutional Review Board (IRB) approval was obtained for this prospective study. Healthy volunteers with any metallic implant considered as a major contra-indication were excluded. All included volunteers gave written informed consent.
Imaging techniques
Imaging was performed on a 7.0 T whole-body MRI system (Philips Healthcare, Cleveland, OH, USA) with a 32-channel receive-coil and a volume transmit/receive coil for transmission (Nova Medical, Inc., Wilmington, MA, USA). The final protocolconsistedofathree-dimensional(3D)magnetizationprepared T1-weighted turbo field-echo (TFE) sequence (duration 8:08 minutes), a 3D T2-weighted turbo spin-echo (TSE) sequence (duration 10:24 minutes), and a 3D T1-weighted magnetization-prepared inversion recovery (MPIR) TSE sequence (duration 10:40 minutes), each with whole-brain coverage. The latter sequence was based on the MPIR-TSE sequence for vessel wall imaging and imaging of the hippocampalanatomy [16–18] This MPIR-TSEsequencewas used because of the low artefacts in the pituitary gland region, which is close to the air in the sphenoid sinus. Total imaging time, including preparation sequences, was approximately 37 minutes. Detailed imaging parameters of all 7.0 T MRI sequences used can be found in Table 1.
Image analysis
Imagesofthe healthyvolunteerswereprocessedonanoffline workstation (Philips). Coronal, axial, and sagittal reconstructions were made of all obtained sequences; care was taken to use identical angulations with all of the sequences for each individual volunteer. Slice thickness was based on the acquired resolution, and no gap was applied. The images of these final 7.0 T protocol sequences were analysed for image quality and pituitary gland coverage. Image quality for 10 healthy volunteers was assessed by two observers (JH and AR). In the case of differences in assessment between the two observers, consensus was reached. For each sequence, 10 items were scored [12], divided into three categories. Each item was evaluated according to a four-point scale: 0 = non-diagnostic, 1 = poor, 2 = moderate, and 3 = good. In the first category, ‘anatomy,’ four items were assessed: a) border between pituitary gland and cavernous sinus, b) border between anterior and posterior pituitary gland, c) visualization of cranial nerves in cavernous sinus, and d) visualization of optic nerve. The second category, ‘artefacts,’ comprised four items: a) susceptibility artefacts, b) pulsation artefacts, c) movement artefacts, and d) B1 inhomogeneity artefacts. In this category, an evaluation of ‘good’ was assigned when there was no loss of image quality due to the specific type of artefact; conversely ‘poor’ indicated diminished image quality due to the specific type of artefact. The last category, ‘image quality,’ consisted of two items: total image quality and diagnostic image for clinical purpose. Mean scores were calculated for these items.
2006 Eur Radiol (2014) 24:2005–2011
Patient evaluation
Additionally, the protocol dedicated for pituitary gland imaging was tested in five patients in whom a clinical 1.5 T MRI of the pituitary gland was not sufficient to diagnose a highly expected microadenoma. A post-contrast T1-weighted TFE and T1-weighted MPIR-TSE sequence after administration of 0.1 mL/kg of a gadoliniumcontaining contrast agent (gadobutrol/Gadovist 1.0 mmol/ mL, Bayer Schering Pharma, Newbury, UK) was added to the protocol in all patients. The 1.5 T and 7.0 T MR images of these five patients were retrospectively evaluated by a single observer (JH). On both field strengths, diagnosis of a visible lesion was based on the combination of sequences. IRB approval was obtained for 7.0 T MRI for clinical purposes in the event that the better differentiating capacity of 7.0 T MRI would lead to more accurate diagnosis. Approval was based on the FDA consideration that a magnetic resonance diagnostic device up to a field strength of 8.0 Tesla is of non-significant risk in adults, children, and infants >1 month of age. (http:// www.fda.gov/downloads/MedicalDevices/ DeviceRegulationandGuidance/GuidanceDocuments/ UCM072688.pdf).
Coverage analysis
For both healthy volunteers and patients, the sequences performed at 7.0 T were analysed for pituitary gland coverage. Theseresultswerecomparedwiththepituitaryglandcoverage on1.5Timagesofthepatientsinordertoassessdifferencesin potential partial volume effects. Coverage was assessed by counting the number of slices covering the pituitary gland, whichwasperformedinallthreedirections.Anincreaseinthe number of slices covering the pituitary gland suggested a lower potential of partial volume effects.
Results
Study population
Themeanageof10includedhealthyvolunteers(4males)was 25 years (range 19–28 years).
Image analysis
Results of quality analysis can be found in Table 2. The T1-weighted MPIR-TSE and T2-weighted TSE sequence scored superior in anatomy, artefacts, and image quality. Figure 1 provides an overview of the 7.0 T MR images obtained in a healthy volunteer, which demonstrate highresolution and highly detailed images with an overall good quality of the T2-weighted TSE sequence and the T1-weighted MPIR-TSE sequence. Due to susceptibility artefacts, the overall image quality of the T1-weighted TFE sequence was poor. This effect was visible in 9 of 10 healthy volunteers.
Patient evaluation
An overview of all patients is provided in Table 3. In four of the five patients, a small hyperintense lesion suggestive of a microadenoma was visible on the 7.0 T images of the pituitary gland (Fig. 2). All four patients underwent surgery. A lesion was found in three of these patients, and was confirmed by histopathological examination to be an adrenocorticotropic hormone (ACTH)-producing microadenoma. The fourth patient was treated in another hospital, where the 7.0 T MR images were not available. Blind surgery yielded no resection of the expected microadenoma, and the patient still suffers from Cushing’s disease. In the absence of follow-up imaging, we are not able to determine whether the visualized lesion is still present.
Table 1 Imaging parameters of the7.0Tprotocol,usedforhealthy volunteers and patients. Acquisition in sagittal direction for each sequence
TFE,turbofieldecho;MPIR-TSE magnetization-preparation inversion recovery turbo spin-echo; T1w, T1-weighted; T2w, T2weighted; TE, echo time; TR, repetition time; TI, inversion time; TSE or TFE factor, echo train length; SENSE factor in anterior posterior(AP)directionand right left (RL) direction
Imaging parameter T1w TFE T 1w MPIR-TSE T 2w TSE
FOV (mm) 200×250×200 250×250×190 250×250×190 Acquired resolution (mm) 1.0×1.0×1.0 0.8×0.8×0.8 0.7×0.7×0.7 Acquired voxel size (mm3) 1.0 0.512 0.343 TR/TI (ms) 8/1200 3952/1375 3200/TE/equivalent TE (ms) 1.97/- 37/19 300/58 Flip angle (degrees) 8 150 120 TFE/TSE-factor 140 158 182 NSA 1 2 2 SENSE factor (AP×RL) 2×3 2×3 2×2.8 Duration (min:sec) 8:08 10:40 10:24
Eur Radiol (2014) 24:2005–2011 2007
Coverage analysis
Results of the pituitary gland coverage are shown in Table 2. Thecoverageat7.0 Twas 16(range13–19),12(range9–15),
and 7 (range 6 –9) slices in sagittal, coronal, and transversal direction, respectively. In comparison, the pituitary gland coverageon1.5TMRIonaveragewas4(range3–5)slicesinsagittal direction and 3 (range 2 –4) slices in coronal direction.
Table 2 Qualitative scores according to 10 items, divided into three categories: anatomy, artefacts, overall image quality. Each itemwas evaluated accordingtoa four-point scale consisting of 0 = non-diagnostic, 1 = poor, 2 = moderate, and 3 = good
TFE,turbofieldecho;MPIR-TSE magnetization-prepared inversion recovery turbo spin-echo; T1w, T1-weighted; T2w, T2-weighted
T1w TFE T1w MPIR-TSE T 2w TSE
Anatomy Border pituitary gland – cavernous sinus 0.2 (0-2) 2.9 (2-3) 3.0 (3) Border anterior – posterior pituitary gland 1.4 (0-3) 3.0 (3) 1.8 (0-3) Visualization cranial nerves 1.3 (0-2) 2.4 (1-3) 1.3 (0-3) Visualization optic nerve 1.6 (0-3) 2.6 (1-3) 2.4 (1-3) Artefacts Susceptibility effects 0.3 (0-2) 2.7 (1-3) 2.6 (1-3) Pulsation artefacts 3.0 (3) 3.0 (3) 3.0 (3) Movement artefacts 2.9 (2-3) 2.8 (1-3) 3.0 (3) B1 inhomogeneity artefacts 3.0 (3) 3.0 (3) 3.0 (3) Overall image quality Total image quality 0.7 (0-3) 2.9 (2-3) 2.8 (2-3) Diagnostic for clinical purpose 0.7 (0-3) 3.0 (3) 2.9 (2-3)
Fig. 1 All obtained 7.0 T MR images of a 26-year-old healthy volunteer, consisting of a T1weighted TFE sequence (a, b, c), T1-weighted MPIR-TSE sequence (d, e, f) and T 2weighted TSE sequence (g, h, i) in sagittal (a, d, g), coronal (b, e, h) and transversal (c, f, i) direction, showing the anterior (white arrowhead) and posterior pituitary gland (open arrowhead) as well as the pituitary stalk (white arrow)
2008 Eur Radiol (2014) 24:2005–2011
Discussion
We have shown that ultra-high-field-strength MRI of 7.0 T is feasible for imaging of the pituitary gland with high spatial resolution. Qualitative good high-resolution images of the
pituitary gland can be obtained with a protocol consisting of aT 1-weighted MPIR-TSE sequence and a T2-weighted TSE sequence. Furthermore, in two patients, lesions that could not be diagnosed on the clinical 1.5 T MRI pituitary images were detected at 7.0 T MRI and confirmed with histopathology. While this study suggests advantages in the use of ultrahigh-field-strength MRI for imaging of the pituitary gland, there are several limiting factors. First, safety test data are limited for metallic implants at 7.0 T. In practice, this restricts the ability to perform 7.0 T MRI in most patients withpostoperative status. However, althoughthe procedure is currently contraindicated in patients with stents and other metallic implants, recent studies confirm that not all metallic implants should be considered as major contraindications for ultrahigh-field-strength MRI [19]. In addition, patients with a pituitary microadenomaare typically young,and the presence of metallic implants is less common.
Table 3 Overview of patients scanned at 7.0 T MRI
Age (years) 1.5 T 7.0 T Surgery PA Remark
35 +/-* + – NA Surgery elsewhere, no cure 45 +/-* – + + 40 +/-* + + + 42 – + + + 14 – + + +
+positivefindings;-negativeornofindings;*non-diagnosticlesionwas visible at 1.5 T MRI
Fig. 2 T2-weighted TSE sequenceandT1-weightedMPIRTSE sequence before and after contrast administration at 7.0 T MRI of a 42-year-old male (a), a 44-year-oldfemale(b),a40-yearold female (c) and a 14-year-old girl with MEN-1 syndrome, (d). All patients suffered from clinical symptoms suspected for Cushing’s disease, and biochemical tests proved an ACTH-producing pituitary adenoma. Arrowheads point to thesuspectedlesionat these7.0T images. Histopathological examination confirmed the lesions of patient A and patient C to be an ACTH-producing microadenoma. The lesion in patient D was confirmed to be a prolactin-producing lesion. A second, smaller lesion in this patient was confirmed to be an ACTH-producing microadenoma
Eur Radiol (2014) 24:2005–2011 2009
Second, issues such as longer T1 values, sensitivity for pituitary enhancement, specific absorption rate (SAR) limitations, magnetic susceptibility artefacts around the air regions (sphenoid sinus) where the pituitary gland is located, and B0/ B1 field heterogeneities are hurdles that must be dealt with at 7.0T,asthepituitaryglandislocateddeepinthemiddleofthe skull and in close proximity to air in the sphenoid sinus. However, the MPIR-TSE and T2-weighted imaging methods that have been developed address these technical constraints and are able to provide images of the pituitary gland with limited amounts of artefacts. Since the pituitary gland is located in the middle of the head, the B0/B1 field heterogeneities werea minor issue with respect tothe image quality of the pituitary gland itself. Susceptibility, meanwhile, was a more disturbing artefact due to the location of the pituitary gland close to the sphenoid sinus. Nevertheless, this effect does not result in overall non-diagnostic image quality of the MPIR-TSE sequence. Another limiting artefact was movement. Due to the high spatial resolution, in combination with the relatively long duration of the sequences, the smallest movement can cause disturbing artefacts. The examination duration of the T1-weighted MPIR-TSE sequence is approximately 10 minutes, which is more than twice that of a standardclinicalT1-weightedTSEsequenceat1.5T.However,in view of partial volume effects, it is important to acquire as many slices covering the pituitary gland as possible. The third limitation is the rather small group of subjects who were evaluated and the lack of comparison with 3.0 T MR imaging, since the current study is a feasibility study of pituitary gland imaging at 7.0 T. While 1.5 T MRI is still the most common in pituitary gland imaging, high-field MRI of 3.0 T has been successfully described [4, 5, 8, 12–14]. These studies demonstrate the superiority of 3.0 T over 1.5 T MR imaging for detection of suspected pituitary gland pathology. Since 3.0 T MRI is more widely available, future studies are warranted in which head-to-head comparisons of 1.5 T, 3.0 T, and7.0T, includingpathologicalconfirmation, are performed in larger groups of patients in order to assess actual clinical added value [4, 5, 8, 12]. Finally, while a lesion visible at 7.0 T MR images in two patientswithCushing’sdiseasewasconfirmedbyhistopathology to be an ACTH-producing microadenoma, the higher spatial resolution of this pituitary gland MRI protocol will also increase the risk of finding incidentalomas. We know from previous studies that incidentalomas in the pituitary gland are present in approximately 20 % of patients [20]. Therefore, it is possible that high-resolution imaging as proposed in the current study may be indicated only for selected patientswithahighsuspicionofpituitarymicroadenomasthat cannot be visualized on lower-field-strength MRI. Interestingly,ontheT1-weightedMPIR-TSEsequence,the pituitary gland lesions visualized in this study had a higher intensity compared to the surrounding pituitary gland after
contrast administration. Classically, on lower field strengths, themajorityofmicroadenomashavealowerintensitythanthe surrounding pituitary gland on standard post-contrast images. There are several theories that may explain this reversedcontrast phenomenon with hyperintense lesions. First, the amount of contrast used was the same as that for lowerfield-strength pituitary MR imaging, and it is possible that a lower dose of contrast is needed when imaging at a higher field strength to produce the same contrast difference [15]. Second, the T1-weighted MPIR-TSE sequence is not purely T1-weighted. A shine-through effect of the contrast agent’s T2* relaxation time may have changed the relative contrast in the images of the pituitary gland lesions. Finally, it is likely thatthewashoutphasehadalreadybeguninthe7.0Tprotocol due to differences in timing of imaging after contrast administration with the long (10: 40 minutes) post-contrast acquisitionoftheT1-weightedMPIR-TSEsequence.Moreextensive research is needed to definitively identify the reason for this phenomenon. Inconclusion,normalanatomyofthepituitaryglandcanbe depicted with a protocol consisting of a T1-weighted MPIRTSEsequenceandaT2-weightedTSEsequenceat7.0TMRI. The protocol dedicated for identification of normal pituitary anatomy also succeeded in identifying micropathology. In future studies, head-to-head comparisons of 7.0 Twith 1.5 T and 3.0 Tare needed with larger samples of patient and with imaging times feasible for clinical settings. However, the current study suggests that high-resolution 7.0 T MRI of the pituitary gland may provide new perspectives when used as a second-line diagnostic examination in the specific context of Cushing’s disease.
Acknowledgements The scientific guarantor of this publication is J. Hendrikse. The authors of this manuscript declare relationships with the following companies: Philips Healthcare, Best, the Netherlands. The authors state that this work has not received any funding. No complex statistical methods were necessary for this paper. Institutional Review Board approval was obtained. Written informed consent was obtained from all subjects (patients) in this study. Methodology: primarily prospective, diagnostic or prognostic study, performed at one institution.
References
1. Escourolle H, Abecassis JP, Bertagna X et al (1993) Comparison of computerized tomography and magnetic resonance imaging for the examination ofthe pituitary gland in patientswith Cushing’s disease. Clin Endocrinol (Oxf) 39:307–313 2. DwyerAJ,FrankJA,DoppmanJLetal(1987)Pituitaryadenomasin patients with Cushing disease: initial experience with Gd-DTPAenhanced MR imaging. Radiology 163:421–426 3. Shah S, Waldman AD, Mehta A (2012) Advances in pituitary imaging technology and future prospects. Best Pract Res Clin Endocrinol Metab 26:35–46
2010 Eur Radiol (2014) 24:2005–2011
4. Kim LJ, Lekovic GP, White WL, Karis J (2007) Preliminary experience with 3-Tesla MRI and Cushing’s disease. Skull Base 17:273– 277 5. Wolfsberger S, Ba-Ssalamah A, Pinker K et al (2004) Application of three-tesla magnetic resonance imaging for diagnosis and surgery of sellar lesions. J Neurosurg 100:278–286 6. PlewesDB,KucharczykW(2012)PhysicsofMRI:aprimer.JMagn Reson Imaging 35:1038–1054 7. Newell-Price J, Trainer P, Besser M, Grossman A (1998) The diagnosis and differential diagnosis of Cushing’s syndrome and pseudoCushing’s states. Endocr Rev 19:647–672 8. Pinker K, Ba-Ssalamah A, Wolfsberger S, Mlynarik V, Knosp E, TrattnigS(2005)Thevalueofhigh-fieldMRI(3T)intheassessment of sellar lesions. Eur J Radiol 54:327–334 9. Runge VM, Patel MC, Baumann SS et al (2006) T1-weighted imagingofthebrainat3teslausinga2-dimensionalspoiledgradientecho technique. Invest Radiol 41:68–75 10. Krautmacher C, Willinek WA, Tschampa HJ et al (2005) Brain tumors: full- and half-dose contrast-enhanced MR imaging at 3.0 T compared with 1.5 T–Initial Experience. Radiology 237: 1014–1019 11. SasakiM,InoueT,TohyamaK,OikawaH,EharaS,OgawaA(2003) High-field MRI of the central nervous system: current approaches to clinical and microscopic imaging. Magn Reson Med Sci 2:133–139
12. Kakite S, Fujii S, Kurosaki M et al (2011) Three-dimensional gradient echo versus spin echo sequence in contrast-enhanced imaging of the pituitary gland at 3 T. Eur J Radiol 79:108–112 13. Erickson D, Erickson B, Watson R et al (2010) 3 Tesla magnetic resonanceimagingwithandwithoutcorticotropinreleasinghormone stimulation for the detection of microadenomas in Cushing’s syndrome. Clin Endocrinol (Oxf) 72:793–799 14. Stobo DB, Lindsay RS, Connell JM, Dunn L, Forbes KP (2011) Initial experience of 3 Tesla versus conventional field strength magnetic resonance imaging of small functioning pituitary tumours. Clin Endocrinol (Oxf) 75:673–677 15. SoherBJ,DaleBM,MerkleEM(2007)AreviewofMRphysics:3T versus 1.5 T. Magn Reson Imaging Clin N Am 15:277–290 16. vanderKolkAG,ZwanenburgJJ,BrundelMetal(2011)Intracranial vessel wall imaging at 7.0-T MRI. Stroke 42:2478–2484 17. van der Kolk AG, Hendrikse J, Brundel M et al (2013) Multisequence whole-brain intracranial vessel wall imaging at 7.0 tesla. Eur Radiol 23:2996–3004 18. Wisse LE, Gerritsen L, Zwanenburg JJ et al (2012) Subfields of the hippocampal formation at 7 T MRI: in vivo volumetric assessment. Neuroimage 61:1043–1049 19. WezelJ,KooijBJ,WebbAG(2013)AssessingtheMRcompatibility of dental retainer wires at 7 Tesla. Magn Reson Med 20. Scangas GA, Laws ER Jr (2013) Pituitary incidentalomas. Pituitary
Eur Radiol (2014) 24:2005–2011 2011

PLACE THIS ORDER OR A SIMILAR ORDER WITH US TODAY AND GET AN AMAZING DISCOUNT