Calculating Susceptibility in High-Resolution Images - Assignment Example

Question1A. MRI of the brain has an excellent contrast when phase images are used in susceptibility weighed MRI. This phase is normally affected by the geometry of the tissues and direction in relation to the magnetic field. In this the changes in phase occur in areas that are even beyond the altered susceptibility (Pruessmann, 2006, p. 299). There are a number of Fourier methods that were developed purposely for calculating susceptibility in high resolution images that are provided in a single resolution. Contrast in cortical layers is usually more consistent in susceptibility images. This contrast was also dependent on the orientation of the structures in relation to B0. In addition the contrast in those images that have iron was more than the phase image contrast. Calculating susceptibility using the iron content can be done more easily by correlating the regions with their iron content. This has an advantage in overcoming the dependency on orientation and non locality of the phase image. Phase in T2 signal has been used to improve the contrast of the image more especially in fields of high magnetic strengths. Scientists have found out that fields of high magnetic strength have high contrast to noise ratio between the white and the grey matter of the brain. This is in comparison of the phase based MRI and those that utilize conventional techniques. It has also been proven that those regions of the brain that are rich in iron there is an improved contrast. Several benefits have been attributed to phase imaging as compared to those images that are of standard magnitude (Yeh et al 2005, p. 1384). This is because they are not dependent on T1 and T2 parameters. It also shows that these phase based images are not so much affected by radiofrequency in homogeneity of high magnetic field strengths. This shows that, phased images could be used for measuring tissue properties. The different tissues and structures have different magnetic susceptibility. These are the differences that cause variations in the magnetic field strength. As a result, there changes that occur in the frequency of the resonance excited and also detected by the MRI. This is what is viewed as the contrast in the phase based images and gradient echo images (Heidemann, et.al 2001 p. 1066. Some of the factors that affect phase include the echo time used, the strength of the magnetic field used, tissue orientation and the structures geometry. One thing that has been discovered is that, the contrast of the grey and the white matters is usually varied in high resolution phase images and appear to vanish. This is considered due to the difference in tissue orientation and the dependency on the orientation. Those phases that are orientation dependent are usually nonlocal (Griswold et.al 2002, p. 1209). The contrast also extends beyond the areas of the susceptibility. Therefore it is true to say that in MRI phase is related indirectly to the susceptibility difference of tissues as well as the difference in susceptibility of air and tissue. This is a hindrance to application of high resolution phase based images that help in studying the anatomy of tissues and their structures. It is therefore important to calculate the susceptibility of tissues because it is related to its composition. Q 1B) In the given case scenario there is need to change the timing parameters. The three fundamental parameters in EPI images are T1, T2 and proton density. Changing the timing parameters influences the signal contrast in tissues and this makes it varied. Increasing the receiver bandwidth for the read out gradient the sensitivity in SE is increased. Consequently it becomes possible to measure quite smaller sizes more theoretically many times thinner. In doing this one can still get the same size artifact but now the restricting factors are spatial resolution and SNR (Kyriakoset. et al, 2000 p. 306). Determining susceptibility is set by the field inhomogeinity that remains after shimming. Therefore, shimming should be done thoroughly and there should always be a prerequisite before a good shim. Question 2. Spiral imaging that is interleaved is in between the EPI methods and the gradient echo. This is relation to the number of radiofrequency pulses that are required. Spiral imaging also has characteristics of benign motion artifacts. It is also suitable for fluoroscopic applications in the trajectory. Circular coils Circular trajectories have an advantage of isotropic resolution and it excludes the K-space corners. Theoretically, it requires much less time for sampling the signal and the resolution of the image is determined by the radial distance as in spiral trajectories. In cardiac imaging, circular EPI has been described as a fly back (Kyriakoset et.al, 2000 p. 308). This is an EPI k-space trajectory that completely eliminates the Nyquist ghosts by obviating echo time reversals. Spiral MRI is stronger to motion artifacts compared to in EPI. Spiral imaging Spiral imaging is more preferred during real time applications which involve dynamic processes. The major challenge associated with spiral during real time applications, is the slow speed in reconstruction. Spiral trajectories do not sample data on rectilinear grids and so the raw data have to be regridded earlier. When calculations on computation costs are done for the spiral reconstruction it is found that the logarithm is too high and slow and as such one requires twenty frames of real time imaging to achieve a minimum speed. Spiral imaging is a form of k-space sampling. It is good in terms of hardware efficiency. Spiral imaging is also fast in imaging and it has robust while flowing (Liu et. al 2008, p. 273). Spiral imaging is mainly used for real time applications that are fast. In this way, spiral trajectories are usually collected after every TR. The spirals rotate in a way that N spirals cover the k-space. In other cases radial sampling can be used to cover k-space sufficiently but this might be considered as private in density spirals. Question 3. Segmented EPI was developed by Mckinnon as a method of utilising EPI on a standard scanner which lacks fast ramping gradients. Segmented EPI has the following Advantages It reduces distortion, improves line width and increases signal to noise ratio. Any irregularities in MRI field while creating a proton density are presented in form of distortion. Even if the field of B0 which is produced by the magnet is homogenous local in homogeneities are caused by the sample that contains regions with different magnetic susceptibility (Kreis, 2004, p. 369). The rate by which the image is affected by the inhomogeinity is dependent on the frequency of the image per point. Hertz separation in the adjacent points is the inverse of sample time length In Fourier transformation, Hertz separation in the adjacent points is the inverse of sample time length. For instance if a signal is sampled to be 100ms the frequency for every point is 10HZ.In Echo planar imaging distortion usually occur in the broadening direction. This is because the mentioned direction has a longer sampling time than the switched direction. This is very different from 2D where distortion usually occurs in read direction (Heidemann, et.al 2001 p. 1066). In 2DFT sampling time is shorter than sample time in EPI in the broadening direction. This is what makes the EPI be more sensitive to the distortion which is in the broadening direction. In case a technique that utilizes two shorts is employed the time spent in sampling each of the FID is greatly reduced to half the time. This result to an increase in frequency by a factor of two, for s every point. The below diagrams are an example of reduction of image distortion while scanning human brain. They utilize the matrix size of 128 x 128 and they are using single shot. The images were obtained by use of a single shot of EPI with two and four shot interleaved EPI. The read gradient was 615HZ and this means that the sampling time for the single, double and the four shot were 104, 52 and 26 ms respectively. Distortion is very apparent in the below figures. It is noted by the increased level of ghosting on the images. (a) (b) (c) The main advantage while using segmented EPI is that when the number of interleaves is increased distortion is reduced. Another advantage of segmented EPI is the Improvement in Line width. This results from the increase of frequency for every point in the image. Segmented EPI has the advantage of increased signal to noise ratio. This can be achieved even without increase in resolution or distortion. It is achieved by lowering the switched frequency gradient resulting to an increase in time taken in sampling a single line in k space. In return the bandwidth required is reduced and signal to noise ratio is increased by the square root of the interleaves used (Griswold et.al, 2000, p. 607). Disadvantages include: Line width broadening can sometimes be a limitation where there are larger matrix and low switching rate gradients. This is where the segmented EPI is used to regain true resolution. A combination of matrix size and gradient rate with number of segment can be beneficial in distortion, resolution signal to noise ratio and gradient heating. However the increase in the number of acquisition may make the experiment take a bit longer than usual. Segmented EPI is a promising approach in achieving high resolution images in MRI. This is because if offers Increase signal to noise ratio which are similar to the traditional 2D EPI readouts. This technique has become so common and is being applied widely (Griswold et.al, 2002, p. 1208). Some of the advantages of this technique include performance of normalization with more keenness than the standard option. However, segment is not very strong on MPRAGE images..This is due to either a large field of view that extends into the neck or the inhomogeinity of the image as a result of parallel imaging. Question 4 In order to scan faster, parallel imaging makes use of multiple coils. This speed in scanning comes from omitting the k-space lines or even spacing the lines apart. In this the field of view is reduced and this produces image warping or even aliasing in case the objects extend beyond the field of view in the encoding direction. SENSE In sensitivity encoding (SENSE) the aliased images are usually reconstructed from each of the coils and later combined by use of coil sensitivity to wrap or cancel the aliasing.. In absence of noise, SENSE is normally mathematically correct. However, it also suffers from amplification. This usually happens in reconstructed images especially so in high reduction factors because of the encoding matrix that is ill conditioned. As a way of solving this problem there have been a number of regularization techniques that have been put into use. There has been a lot f challenges in minimising scan time. This is because the image that is prior is always at low resolution and this has the possibility of resulting to blurring or even artifacts for the high rate of accelerating in reconstruction. In SENSE reconstruction is done in the space of the image utilizing the coil sensitivity which had been acquired previously in coil sensitivity (Barker & Lin, 2006, p.116). This means that SENSE process of reconstruction can only be done after the raw data is sampled. SMASH In SMASH reconstructs the missing k space uses data of linear combination so that the interleaved data and reconstruction of the K space is made possible. In case of SMASH the k-space that is missing is restored by utilizing sensitivity that comes before reconstruction (Blaimer et.al, 2004 p. 227). Comparison of SMASH and SENSE Nevertheless in both SMASH and SENSE the missing data is reconstructed by the algorithms without artifacts that are back folding. The difference in the two lies in how the way this is actually done. What happens is that SMASH performs the calculation on raw data previous to Fourier transform. On the other hand SENSE will use images formed from the coils. The two techniques are utilised on the basis of knowledge of profiles sensitivity of every coil. The mentioned knowledge is acquired in two ways. One, it is either through a 3D volume scan that is of very low resolution (Buades & Morel, 2008, p.126). This can then be used for the subsequent scans that are independent of the orientation. It can or is acquired by adding phase encoding steps on each reduced acquisition. This means that the first solution is effective if more scans are to be performed while the latter is preferred where only one scan is needed. Bibliography: Barker PB, Lin DDM. (2006)In vivo proton MR spectroscopy of the human brain. Prog Nucl Magn Reson Spectros vol (49):99-128. Buades A, Coll B, Morel JM.(2008) Nonlocal image and movie denoising. Int J Comput Vision vol;76:123–139. Blaimer M, Breuer F, Mueller M, Heidemann RM, Griswold MA, Jakob PM. (2004) SMASH, SENSE, PILS, GRAPPA: how to choose the optimal method. Top Magn Reson Imaging. Vol ;15(4):223–36 Griswold MA, Jakob PM, Heidemann RM, Nittka M, Jellus V, Jianmin W, Kiefer B, Haase A.(2002) Generalized autocalibrating partially parallel acquisition. Magn Reson Med.vol ;47(6):1202–1210. Griswold MA, Jakob PM, Nittka M, Goldfarb JW, Haase A.(2000) Partially parallel imaging with localized sensitivities (PILS) Magn Reson Med. vol; 44(6):602–609. Heidemann RM, Ozsarlak O, Parizel PM, Michiels J, Kiefer B, Jellus V, Muller M, Breuer F, Blaimer M, Griswold MA, Jakob PM.(2003) A brief review of parallel magnetic resonance imaging. Eur Radiol. Vol;13(6):2323–2337 Heidemann RM, Griswold MA, Haase A, Jakob PM. (2001) VD-AUTO-SMASH imaging. Magn Reson Med. 2001;45(6):1066–1074. Kreis R. (2004) Issues of spectral quality in clinical 1H-magnetic resonance spectroscopy and a gallery of artifacts. NMR Biomed 2004;17(6):361-381. Kyriakos WE, Panych LP, Kacher DF, Westin CF, Bao SM, Mulkern RV, Jolesz FA.(2000) Sensitivity profiles from an array of coils for encoding and reconstruction in parallel (SpaceRIP) Magn Reson Med. vol;44:301–308. Liu YL, Wang J, Chen X, Guo YW, Peng QS.(2008) A robust and fast nonlocal means algorithm for image denoising. J Comput Sci Technol vol;23:270–279 Pruessmann KP.(2006) Encoding and reconstruction in parallel MRI. NMR Biomed. 2006;19 (3):288–299. Yeh EN, McKenzie CA, Ohliger MA, Sodickson DK. (2005) Parallel magnetic resonance imaging with adaptive radius in k-space (PARS): constrained image reconstruction using k-space locality in radiofrequency coil encoded data. Magn Reson Med. vol; 53(6):1383–1392. Read More

This is a hindrance to application of high resolution phase based images that help in studying the anatomy of tissues and their structures. It is therefore important to calculate the susceptibility of tissues because it is related to its composition. Q 1B) In the given case scenario there is need to change the timing parameters. The three fundamental parameters in EPI images are T1, T2 and proton density. Changing the timing parameters influences the signal contrast in tissues and this makes it varied.

Increasing the receiver bandwidth for the read out gradient the sensitivity in SE is increased. Consequently it becomes possible to measure quite smaller sizes more theoretically many times thinner. In doing this one can still get the same size artifact but now the restricting factors are spatial resolution and SNR (Kyriakoset. et al, 2000 p. 306). Determining susceptibility is set by the field inhomogeinity that remains after shimming. Therefore, shimming should be done thoroughly and there should always be a prerequisite before a good shim.

Question 2. Spiral imaging that is interleaved is in between the EPI methods and the gradient echo. This is relation to the number of radiofrequency pulses that are required. Spiral imaging also has characteristics of benign motion artifacts. It is also suitable for fluoroscopic applications in the trajectory. Circular coils Circular trajectories have an advantage of isotropic resolution and it excludes the K-space corners. Theoretically, it requires much less time for sampling the signal and the resolution of the image is determined by the radial distance as in spiral trajectories.

In cardiac imaging, circular EPI has been described as a fly back (Kyriakoset et.al, 2000 p. 308). This is an EPI k-space trajectory that completely eliminates the Nyquist ghosts by obviating echo time reversals. Spiral MRI is stronger to motion artifacts compared to in EPI. Spiral imaging Spiral imaging is more preferred during real time applications which involve dynamic processes. The major challenge associated with spiral during real time applications, is the slow speed in reconstruction.

Spiral trajectories do not sample data on rectilinear grids and so the raw data have to be regridded earlier. When calculations on computation costs are done for the spiral reconstruction it is found that the logarithm is too high and slow and as such one requires twenty frames of real time imaging to achieve a minimum speed. Spiral imaging is a form of k-space sampling. It is good in terms of hardware efficiency. Spiral imaging is also fast in imaging and it has robust while flowing (Liu et.

al 2008, p. 273). Spiral imaging is mainly used for real time applications that are fast. In this way, spiral trajectories are usually collected after every TR. The spirals rotate in a way that N spirals cover the k-space. In other cases radial sampling can be used to cover k-space sufficiently but this might be considered as private in density spirals. Question 3. Segmented EPI was developed by Mckinnon as a method of utilising EPI on a standard scanner which lacks fast ramping gradients. Segmented EPI has the following Advantages It reduces distortion, improves line width and increases signal to noise ratio.

Any irregularities in MRI field while creating a proton density are presented in form of distortion. Even if the field of B0 which is produced by the magnet is homogenous local in homogeneities are caused by the sample that contains regions with different magnetic susceptibility (Kreis, 2004, p. 369). The rate by which the image is affected by the inhomogeinity is dependent on the frequency of the image per point. Hertz separation in the adjacent points is the inverse of sample time length In Fourier transformation, Hertz separation in the adjacent points is the inverse of sample time length.

For instance if a signal is sampled to be 100ms the frequency for every point is 10HZ.In Echo planar imaging distortion usually occur in the broadening direction.

Read More