Advantages of Higher-Field Tesla Scanners - Essay Example

Tesla scanners Introduction The quantity of available signal in conventional magnetic resonance imaging (MRI) is inevitably linked to the static magnetic field strength of the magnetic system. In the recent past clinical MRI strengths operated at fields strengths at or below 1.5 Tesla. However, due to improvements in the magnetic design and shielding to ease the sitting requirements, 3 Tesla clinical scanners have been improvised and implemented and are now widely available (Gaa et al, 2004). The push for high-field MRI is wholly tied to the benefits of potentially higher signal-to-noise ration, contrast and spectral resolution for certain specific applications. These benefits often facilitate higher spatial and/or temporal resolutions as compared to the previous models. However, there are various challenges that must be resolved in order to achieve the perceived benefits of 3T scanners. This paper is therefore focused to discuss the advantages of going to higher field MRI. This paper will also discuss the challenges of going to higher field. The influence of the 3T magnet on image quality a. Advantages The introduction of 3T scanners in clinical applications has completely changed the way body diagnosis is being done across hospitals the world over. Faster scans, reduced image degradation, increased contrast between tissues, sub-millimetre resolutions and much more can be captured with a short period. Extensive discussion about the benefits of 3T scanners is as follows. Imaging at 3Tb has been greatly enhanced by various factors. One of the factors that influence imaging at 3T magnets is the radio frequency waves (RF). In MR imaging the localization is limited to spatial resolution lower than the wavelength. As a consequence localization of imaging is encoded in resonance frequency. The sounds emitted by identical body voxels are indistinguishable (Fischbach et al 2004). However, because of the varying liquid content of various body voxels allows distinction of the frequencies of sounds produced by these body voxels. The RF wavelength (λ) is proportional to the speed of light in the medium C medium, and inversely proportional to RF frequency f: λ α The varying fluid content of body voxels causes variation in radio frequency which consequently allows users to acquire images with spatial resolutions and greater coverage than in the case of lower field scanners (Fschbach et al, 2004). Owing to the fact that resonance frequency of a specified voxel is directly proportional to the magnetic field strength, one can obtain a spatial variation of resonance frequency by varyiong magnetic field strength through varying the gradient. Since 3T scanners permits variation of the gradient by taking images from various positions of the patient, one can be able to vary magnetic field strength based on variation of field strength and hence to obtain variation in spatial resonance frequency. This permits up to 181 cm coverage, which makes it more essential in musculoskeletal studies in a single scan without reconfiguring the coils or changing the patient’s position (Fischbach et al 2004). It is impossible for one to localize signals from various directions at the same time hence in most cases imaging method employed at 3T scanners is 2 dimension Fourier Transform (2DFT) imaging. This method applies three gradients successfully that employs three different localization principles: selective excitation, frequency encoding and phase encoding unlike previous MRI scanners which could not integrate the three gradients. Thus the image obtained at 3T scanners is of a higher contrast as opposed to those obtained from previous scanners. The image below illustrates how images are obtained from various gradients to obtain a clear image (Lutterbey et al, 2005). The Gx and Gz gradients are applied simultaneously and result in a unique gradient (Gr) which is a vector sum of the two components. Another factor that impact on the image obtained at 3T scanners is the gradient applied during the excitation pulse. This gradient permits slice selection. In MR the excitation lacks a unique frequency but it instead has a frequency band. Thus, in presence of a gradient, voxels with characteristic frequency are the only ones included in the spectrum of the excitation pulse resonate (Lutterbey et al, 2005). This makes it possible to flip their magnetization in a plane perpendicular to the magnetic field as shown in the diagram below. The selective excitation corresponds to the application of a gradient during the excitation pulse. Since volume elements of equal resonance frequency are located in planes perpendicular to the gradient direction, the excited volume corresponds to a single slice. This implies that the images obtained at 3T scanners have ultra high resolutions that allows distinction and resolution of micro elements such as micro-tumours that were otherwise invisible using previous scanners. The applied gradient during acquisition allows frequency encoding. For liner objects, application of a gradient after a 90o pulse allows establishment of a linear relationship between each voxel and its position along the gradient direction (Lutterbey et al, 2005). The signal received enables the measurement of magnetization in each voxel after Fourier transform (FT). Thus frequency encoding corresponds to the gradient applied during acquisition as shown ion the diagram below. Since the 3T scanners allow application of gradient in various positions, various images can be acquired to provide images of high resolution (Drangova & Pelc, 2001). The magnetic moments of different voxels are dephased progressively as gradient is applied during acquisition. Stepwise increasing of magnetic field gradient which defines phase encoding can also result in progressive dephasing of the magnetic moments of different voxels. Since 3T scanners have a wider magnetic field strength, phase encoding is well defined. This provides well resolved images as shown in the diagram below. MR system is not capable of directly acquiring image but it rather gets a matrix of the numerous elementary signals gathered during the image acquisition. The axes of this matrix are kx and ky and the matrix is called k-space (Drangova & Pelc, 2001). The k-space has similar information as the image but not in a spatial order. K-space is a representation of spatial frequencies in the image which as discussed above very enhanced at 3T scanners. The k-space center has information on slow variations of contrast while the periphery has information on the finest details. Given that spatial frequencies are enhancved at 3T scan the k-space more information on finest details as opposed slow variation information. As a consequence, the images obtained using 3T scanners are more detailed than those obtained using porevious scanners. The diagram below is a representation of k-space and image in MRI (Uematsu et al, 2004). . The image in MR system is obtained by the Fourier transform called k-space instead of the MR imager. The image acquisition entails k-space scanning. The k-space scanning results from the mathematical analysis of the evolution of magnetic moments in the presence of gradients (Uematsu et al, 2004). MR acquisition can be illustrated by the point by point acquisition of k-space as shown in the diagram below. The gradients and pulses allow the k-space to move in various directions. Since at 3T scanners one can use various gradients several images can be obtained that provide high resolution images. The excitation pulses places k-spaceship at point O center of k-space. The velocity of k-spaceship in the directions kx and ky is proportional to gradient intensity Gx and Gy respectively. A 180o pulse takes k-spaceship from a point to its symmetric relative to center O. each output signal from the receiver coil represents a line of k-space whose position is determined by the pulse and gradient sequence (Uematsu et al, 2004). Based on the fact that pulse and gradient is enhanced at 3T imaging, the images obtained are of high resolution. Image acquisition in MR can be accelerated through the use of rough localization provided by surface coils in parallel imaging. Non-uniform images of anatomical structures are obtained when the signal intensity vary with distance which is caused by surface coils. The 3T scanners have however transformed this drawback into an advantage for localization of signal since the sensitivity map of the coils contains spatial information (Uematsu et al, 2004). This information is integrated in the reconstruction of the image after simultaneous acquisition of the same object by multiple surface coils, in order to reduce the number of signals to acquire and therefore it shortens acquisition time at 3T scanners. The diagram below illustrates the effect of coils on image. b. Challenges However, regardless of the above advantages of 3T scanners it is too soon significantly comment on the patient outcomes. For instance, in order to have an all friendly environment at 3T, then the following challenges must be addressed. Patient safety is one of the most serious concerns that 3T scanners have raised. Increase in the main field causes ferromagnetic projectiles, increased torques on medical devices and implants, and increases electromagnetic/magneto-hydrodynamic effects (Drangova & Pelc, 2001). At 1.5T scanners medical devices declared MR safe but this is not at higher fields. The torque on the ferrous object increases linearly with field strengths while the translational force experienced by a ferrous object increases as a product of B0(∆B0). Consequently, increased translational force with regard to heavily shielded units will be larger because of large spatial gradient of the magnetic field outside the scanner (Lutterbey et al, 2005). Although there is no health hazard reported with regard to the use of 3T scanners, it is important to note that there are increased magneto-hydrodynamic effects and human response to high fields. Patients reported the following side effects after using 3T scanners: nausea and vertigo, headache, tingling, visual disturbances and pain associated with tooth fillings. With 3T scanners electrically conductive fluid flow in a magnetic field, this induces an electric current and a force that opposes the fluid flow; particularly when the flow is perpendicular to the field the effects are high. This has a T-wave swelling consequence which is seen on the ECG especially during highest flow through the aorta during an MRI exam. In case of induced potentials of 5 mV/Tesla order will automatically be exacerbated at high fields making it difficult to obtain good ECG’s in the MR environment (Uematsu et al, 2004). Radiofrequency fields are other technical challenges in relation to the use of 3 Tesla and above. As it has been established, B1 field sensitivity increases linearly, but changes increasingly to inhomogeneous due to permittivity, conductivity, and conformation of the patient. The RF poser deposition is usually measured by SAR. In the clinically used field strengths, 0.2T < B0< 3.0T, SAR scales quadratically with field strengths: SAR α B02 With regard to this, the dielectric nature of tissues gives rise to a region of hypersensitivity especially at the field focusing (centre of the image) (Gibbs et al, 2004). Transmission line (TEM) was designed to reduce or compensate the effects. However, post-processing methods based on signal sensitivity variation across the image is a significant challenge at high fields (Uematsu et al, 2004). At high field, the SAR limitations are responsible for creating difficult challenges by forcing tradeoffs between image acquisition rates, resolution and slice coverage. This is specifically evident with high SAR sequences like fast spin echo, chemical fat saturation or magnetization transfer sequence. Furthermore, pulse sequence design and RF pulse design are some of the technical challenges that demand for the acquisition of new hardware designs to overcome these limitations. This implies that extra costs have to be incurred to employ new techniques such as reduced flip angles or using SENSE (Stafford, 2006). Higher gradient fields for imaging and shimming are necessary and desirable at any field strength. However, at higher field strengths, higher performance gradients are compulsory in order to take advantage of higher resolution scans available via the increased SNR from the main field. Despite the value of higher performance gradients, the challenge is the physiological constraint on dB/dt for safety that has an ultimate limit on gradient performance in higher-field systems. In addition, acoustic noise and concomitant field effects are other technical challenges directly associated with higher field strengths (Gibbs et al, 2004). Contrast changes at higher fields and application. Imager signal to noise ration and transmission in high field range is linear dependent. Higher resolutions scans can be performed over a short time span, faster imaging techniques can also be used with similar resolutions or a combination of both (Stafford, 2006). However, relaxation parameters change with increased field strength which makes the use of increase SNR difficult. The longitudinal relaxation time (T1) increases with increase in B0 in tissues like brain parenchyma. For instance, spin-lattice relaxation time, T1, increases and converges for tissues, thus decreasing both SNR and contrast for T1-weighted exams using TR used at 3 Tesla. This causes longer pulse repetition. Consequently, the loss of contrast in T1 weighted images implies that inversion recovery should be implemented in order to achieve the desired tissue contrast between tissues (Stafford, 2006). Spin-spin relaxation, T2, shortens at higher fields. This results into the susceptibility gradients that lead to T2* shortening which increases linearly with the main field causing a shortening of T2* at higher fields. This culminates into better T2* contrast which advantageous in applications such as dynamic susceptibility imaging for profusion (Stafford, 2006). However, shortened T2* values lead to signal loss especially for long TE gradient echo acquisition. This is a big challenge for echo-train based techniques like the echo-planar imaging (EPI). The equation that is associated with this scenario also referred to as experimental relation rate is given by: 1/T2* (1/T2*=1/T2 + 1/T2 (inhomogeneity)) In conclusion, form the above discussion it is true that higher field promises higher SNR and contrast changes that are beneficial to some applications while making others difficult. As technology of image quality advances, high field scanners are the only option to extensively exploit this field of medicine. Regardless of their advantages, 3Tesla scanners have considerable number of technical challenges that must be ironed out before the scanners can be comprehensively and fully used as the 1.5 Tesla scanners. For instance, the desired workflow and routine scan protocol should seriously be evaluated and considered before deciding on high-field as many of the imaging protocols are still in progress. In addition a variety of coils and technology standards are not available on the 3T scanners. Bibliography Fischbach, F., Thormann, M., and Ricke J. (2004) 1H magnetic resonance spectroscopy (MRS) of the liver and hepatic malignant tumors at 3.0 Tesla [in German]. Radiologe, vol. 44, pp. 1192-1196. Lutterbey, G., Gieseke, J., von Falkenhausen, M., Morakkabati, N., and Schild H. 2005. Lung MRI at 3.0 T: a comparison of helical CT and high-field MRI in the detection of diffuse lung disease. Eur Radiol, pp. 324-328. Stafford R. J. 2006. High Field MRI: Technology, Applications, Safety, and Limitations. The University of Texas M. D. Anderson Cancer Center, Houston, TX Drangova, M., and Pelc, NJ. 2001. Artifacts and signal loss due to flow in the presence of B(o) in homogeneity. Magn Reson Med, pp. 126-30. Uematsu H, Takahashi M, Dougherty L, Hatabu H. 2004. High field body MR imaging: preliminary experiences. Clin Imaging, pp. 159-62 Gibbs G.F., Huston J 3rd, Bernstein M.A., Riederer S.J, Brown RD Jr. 2004. Improved image quality of intracranial aneurysms: 3.0-T versus 1.5-T time-of-flight MR angiography. AJNR Am J Neuroradiol, pp. 84-7. Gaa J, Weidauer S, Requardt M, Kiefer B, Lanfermann H, Zanella FE. 2004. Comparison of intracranial 3D-ToF-MRA with and without parallel acquisition techniques at 1.5T and 3.0T: preliminary results. Acta Radiol., pp. 327-32. Read More