BR102017019121A2 - DEVICE FOR THE QUALITY CONTROL OF DIFFUSION TENSOR IMAGES, DIFFUSION-WEIGHT IMAGES FOR RECONSTRUCTION OF TREAT AND FIBER IMAGES AND OTHER PARAMETERS IN MAGNETIC RESONANCE - Google Patents

Abstract

The present invention is in the technical field of quality control and calibration of magnetic resonance imaging equipment and in quality control and calibration applicable to diffusion tensor (dti) and diffusion-weighted image acquisition techniques. (dwi) for tract and fiber reconstruction.

Description

DEVICE FOR THE QUALITY CONTROL OF DIFFUSION TENSOR IMAGES, WEIGHTED DIFFUSION IMAGES FOR RECONSTRUCTION OF TREATMENT AND FIBER IMAGES AND OTHER PARAMETERS IN MAGNETIC RESONANCE

FIELD OF THE INVENTION:

[001] The present invention falls within the technical field of quality control and calibration of magnetic resonance imaging (RM) acquisition equipment and in quality control and calibration applicable to diffusion tensor (DTI) image acquisition techniques. and diffusion-weighted images (DWI) for tract and fiber reconstruction.

[002] The proposed invention aims to solve the technical problem of the lack of a Phantom (simulator, or phantom, in Portuguese) and a quality control protocol (QC) to evaluate the quality of the data obtained in magnetic resonance images weighted by diffusion (DW or DWI images) and diffusion tensor images (DT or DTI images). Additionally, the device allows the evaluation of other quality parameters in MRI, such as: slice thickness, signal-to-noise ratio (SNR), spatial resolution and detection of geometric distortions, as well as the QC in DWI acquisitions that aim to reconstruct tracts and fibers of biological tissues of humans, animals, vegetables and even objects.

TECHNICAL STATUS:

[003] It is known that magnetic resonance images weighted by diffusion or simply DWI, reveal microscopic aspects related to tissue physiology, based on the diffusion characteristics of water molecules

Petition 870170066292, of September 6, 2017, p. 5/67

2/35 through them. In biological tissues, the diffusion of water molecules is isotropic, with the same magnitude in all directions or anisotropic, following a preferred direction. The acquisition of DWI involves the use, in pulse sequences, of diffusion weighting gradients.

[004] In this respect, it is important to note that DWI was established in clinical practice for the diagnosis of stroke in the acute phase and to classify it as ischemic or non-ischemic. From these images it is possible to calculate the apparent diffusion coefficient (ADC) maps and also the diffusion tensor. ADC maps have this name because the value measured for this coefficient is lower than the value measured for free water, diffusing without the presence of barriers and cell compartments. The calculation of these maps is based on the DWI and the T2-weighted image, as shown in Equation 1.1:

Ç·

ADCt = - - ln - (1.1) b S _o

Where b is the diffusion sensitization factor in s / mm ² . It should be noted that in brain studies, the values of b used typically vary between 700 s / mm ² and 800 s / mm ² for children and for adults the most used value is 1000 s / mm ² . This is because Si is the DWI obtained from diffusion gradients in the direction ie S0 is a T2-weighted image. In diffusion pulse sequences, the echo time (TE) must be long to accommodate the duration of the diffusion gradients. Consequently, DWI are based on T2 weighting. In DWI, areas with a high diffusion index tend to

Petition 870170066292, of September 6, 2017, p. 6/67

3/35 are hypointense and areas with a low diffusion index tend to be hyperintense.

[005] In the tensor model, on which DTI, or DT images, are based, diffusion is described as a multivariate Gaussian distribution, with the tensor being a 3x3 covariance matrix, as shown below:

/ D _xx D _X y D _xz \ T = (Dy _X Dyy Dy _Z ), (1.2 \ D _ZX D _Z y D _ZZ J

[006] This matrix describes the displacements of water molecules in 3 dimensions normalized by the diffusion time. The diagonal elements are the diffusion variances along the x, y and z axes and the elements outside the diagonal are the symmetric covariance terms. Given the symmetry of the tensor, for its construction it is necessary to acquire DWI in at least 6 directions to obtain the six independent components. However, to increase the SNR, in most applications the acquisition of DWI is used in 16, 32 or 64 directions. The tensor can be viewed as an image, this technique being called DTI (Diffusion Tensor Imaging). From the images of the diffusion tensor (also called DTI) one can extract scalar parameters (indices) such as fractional anisotropy (FA), a measure of the fraction of the tensor that is anisotropic; the relative anisotropy (RA), the relationship between the magnitudes of the isotropic and anisotropic portions of the tensor; the mean diffusivity (MD), mean of the eigenvalues from the tensor diagonalization; and the volume ratio (VR), a measure that represents the ratio between

Petition 870170066292, of September 6, 2017, p. 7/67

4/35 the volume of an ellipsoid and the volume of a sphere, ranging from 1 (isotropic diffusion) to 0 (anisotropic diffusion). DTIs are also used to reconstruct images of axonal fibers, skeletal muscle fibers, cardiac and smooth muscles, using a technique called tractography. This technique can also be applied to the characterization of objects that have tracts and fibers.

[007] Currently, it is known that DTI and its parameters are quite sensitive to factors related to image acquisition and processing. The reliability of the results obtained from these images depends on the quality of the acquired data. Low SNR, patient movement during the examination, chemical displacement, inhomogeneity of the B0 magnetic field (the main magnetic field of MRI), radio frequency (RF) interference, parasitic currents and magnetic susceptibility effects can degrade the quality of DWI and DTI. Standard QC phantoms for MRI scanners, have diffusivities and relaxation parameters different from biological tissues and are isotropic. They can be useful for the evaluation of parameters such as B0 homogeneity, SNR, contrast-to-noise ratio (CNR), detection of low contrast structure, uniformity, presence of geometric distortions, cut thickness and the spatial resolution of these images, but not for the DTI QC. In addition, DWI and DTI are more sensitive to degradation compared to conventional MRI images and therefore data corruption occurs first in these images. In other words, when a problem is detected in T1, T2 and proton density (PD) images, the DWI and DTI have probably been degraded.

Petition 870170066292, of September 6, 2017, p. 8/67

5/35 [008] When dealing more specifically with phantoms for MRI QC, document CN102334991 describes a comprehensive test phantom for DTI QC. The device comprises a housing that has a test layer and is filled with a test solution. This phantom model also comprises a region containing bundles of fibers, a test region that simulates the human brain and a region for basic QC tests of MRI images. The region destined for the evaluation of DTI uses silicone and cellulose fibers that make it unstable over time, thus making it difficult to monitor the behavior of the MRI scanner. In addition, the presence of several spiral structures contributes to the occurrence of magnetic susceptibility effects.

[009] Still, as an example, it is possible to mention another phantom that allows to measure the anisotropic diffusion, the diffusion of the main axis, the diffusion route of the main axis and to evaluate the effectiveness of DTI, as revealed in the document US7667458. However, such a device has anisotropic structures fixed in specific positions and spiral arrangements that make it more sensitive to the effects of magnetic susceptibility. In addition, the use of microtubes based on organic compounds makes the device not stable in the long term, since they undergo biological degradation.

[010] US7521931 describes a phantom equipped with a reservoir that comprises a plurality of structures that simulate the anisotropy of biological tissues. The document mentions the use of fibers for the formation of these structures. However, in the aforementioned phantom, the

Petition 870170066292, of September 6, 2017, p. 9/67

6/35 mounted microtubes are fixed to the container wall by adhesives, which tends to lead to the formation of air bubbles and consequently to the generation of magnetic susceptibility artifacts, which can also be caused due to the spiral structure in the center of the phantom. Additionally, it is not a device of great stability in the long term, since it contains organic materials and other compounds of plant origin.

[011] Yet another document, such as US4829839, describes an anisotropic diffusion phantom and a method for calibrating diffusion tensor images. The phantom comprises thin plates of glass and glass microtubes among which water diffuses. However, such microscopic structures are more subject to magnetic susceptibility, which can lead to degradation of the obtained data and / or imperfect calibrations.

[012] A more complex construction equipment that allows the assessment of the cut thickness, SNR, geometric distortions and spatial resolution, was described in document US5036280 that reveals a phantom that presents a pattern of holes, capable of evaluating the cut thickness, it is also equipped with several disks with specific structures for evaluating different parameters. However, the complexity in manufacturing, and the great proximity between the different discs can lead to the occurrence of artifacts in the images. In addition, the phantom proposed in the US5036280 patent makes no mention of the possibility of evaluating DWI and DTI acquisitions.

[013] Regarding the DTI QC, the article Wang et al., 2011 - A quality assurance protocol for diffusion tensor

Petition 870170066292, of September 6, 2017, p. 10/67

7/35 imaging using the head phantom from American College of Radiology ”, describes a procedure to ensure the quality of DTI using a phantom developed by the American College of Radiology (ACR). However, the phantom used in this article does not have specific modules that allow the simulation of isotropic or anisotropic diffusion for the real evaluation of the quality of this type of images and does not have elements with parameters of isotropic diffusion, similar to those found in biological tissues.

[014] The phantom described in the article by Teh et al., 2015 - Bíomímetíc Phantom for Cardiac Diffuslon MRI ”, simulates the anisotropic diffusion of water molecules in cardiac muscle cells. The phantom comprises cylindrical microfiber layers made up of poly-polyolactone and polyethylene oxide in continuous bands, with the fibers in each layer being held together by a 20 pm thick Dyneema® fiber filament. However, the proposed phantom is difficult to manufacture, since depending on the variation of the voltage applied in the production, beams of different thicknesses can be generated. In addition, it has low assembly stability, usually 4 months, making it difficult to set the parameters of the MR scanner over long periods of time using the same device. Finally, microstructures such as those used in their manufacture are more subject to the effects of magnetic susceptibility and partial volume, which can alter the results obtained.

[015] Finally, the article by Fieremans et al.,

2008 “Simulation and experimental verification of the diffusion

Petition 870170066292, of September 6, 2017, p. 11/67

8/35 in an anisotropic fiber phantom ”, describes a phantom composed of hydrophobic and multifilament fibers, which is favorable for simulating the anisotropy of biological tissues. The fibers, having a thickness of 20 pm, are laid out in parallel and compacted inside a polyolefin tube. However, it can be said that the use of low density fiber bundles causes them to agglomerate in the center of the heat shrink tube, which, in addition to causing variations in measured DTI parameters, can lead to a greater occurrence of magnetic susceptibility in the images.

[016] It can be seen, therefore, that the current state of the art lacks solutions that allow, considering the diversity of available methods (diffusion, perfusion, functional images, etc.), to verify the quality of different MRI techniques, using the same device.

OBJECTIVES AND ADVANTAGES OF THE INVENTION:

[017] In view of the above, the present invention basically aims to solve the technical problem of the lack of a device (phantom) or QC protocol that allows: i) the quality evaluation of the data obtained from weighted magnetic resonance images by diffusion, DWI, and images of the diffusion tensor, DTI, ii) the evaluation of section thickness, SNR, spatial resolution and detection of geometric distortions and iii) the evaluation of the quality of DTI for reconstruction of tracts and tissue fibers of humans, animals, vegetables and even objects.

Therefore, it is one of the objectives of the present invention to reveal a device for the QC of diffusion tensor images, of diffusion-weighted images for

Petition 870170066292, of September 6, 2017, p. 12/67

9/35 reconstruction of tract and fiber images and other MRI parameters.

[018] It is, therefore, an objective of the present invention to enable the evaluation of the data listed above in a single device, without requiring the availability, acquisition and / or installation of a plurality of different devices and, therefore, making it possible to obtain a QC for images in a simple and practical way.

BRIEF DESCRIPTION OF THE INVENTION:

[019] The aforementioned objectives are achieved through the device for the quality control of images of the diffusion tensor, diffusion-weighted images for reconstruction of images of tracts and fibers and other parameters in MRI, the device comprising at least one cylinder, a first module disposed in the cylinder, a second module disposed in the cylinder, a third module disposed in the cylinder and cooperating with the first and the second modules, at least one spindle for the cooperation of the first, second and third modules with the cylinder, with a fourth module disposed in the cylinder, the said device also comprising, a cooperating cover with the cylinder for fixing the spindles and a liquid solution to allow the diffusion of the water molecules.

BRIEF DESCRIPTION OF THE FIGURES:

[020] To obtain total and complete visualization of the object of this invention, figures are presented which are referred to, as follows.

[021] Figure 1.1 is a three-dimensional representation of one of the possible configurations of the device for image quality control of the diffusion tensor, of diffusion-weighted images for image reconstruction

Petition 870170066292, of September 6, 2017, p. 13/67

10/35 tracts and fibers and other parameters in magnetic resonance 100. This configuration includes all the modules described in this document;

[022] Figure 1.2 is a three-dimensional representation of one of the possible configurations of device 100 shown in Figure 1.1. In this configuration, only the module for evaluating geometric distortions 20 (with circular holes 21) is represented;

[023] Figure 1.3 is a three-dimensional representation of one of the possible configurations of device 100 shown in Figure 1.1. In this configuration, only the module for evaluating geometric distortions 20 (with square holes 21) is represented;

[024] Figure 1.4 is a three-dimensional representation of one of the possible configurations of device 100 shown in Figure 1.1. In this configuration, only the module for evaluating the slice thickness 30 and the module for evaluating the spatial resolution 50 are represented;

[025] Figure 1.5 is a view of the representation illustrated in Figure 1.4. In this representation, the section thickness module 30 was opened, so that blocks 51, 52 and 53 for the assessment of spatial resolution can be viewed;

[026] Figure 1.6 is a three-dimensional representation of one of the possible configurations of device 100 shown in Figure 1.1. In this configuration, only the DTI 40 QC module is represented;

[027] Figure 1.7 is a side view of the main container 101 (cylinder) of the device described in Figure 1.1.

[028] Figure 1.8 is a representation of the top view

Petition 870170066292, of September 6, 2017, p. 14/67

11/35 of one of the possible configurations of the lid 110 of the container 101 (cylinder) of the device shown in Figure 1.1. In this representation, the corresponding dimensions of the cover 110 of the device are detailed.

[029] Figure 1.9 (A) is a representation in top view of a possible configuration of the base 116 in which the lid of the container 110 (cylinder) is fixed, highlighting its corresponding dimensions. Figure 1.9 (B), is a representation of the side view of the structure represented in Figure 1.9 (A). It highlights the possible dimensions of cavity 121 into which the O'ring 115 sealing ring is inserted, which prevents liquid leaks from inside the container.

[030] Figure 1.10 is a two-dimensional representation isolated from the module for evaluating geometric distortion 20. Two possible configurations of this module are presented, in this case consisting of an acrylic plate containing 21 circular holes (1.10 (A)) or 21 square holes (1.10 (B)).

[031] Figure 1.11 is an isolated two-dimensional representation of the plates that make up the section thickness module 30 in one of its possible configurations. In 1.11 (A), 1.11 (B) and 1.11 (C) are the dimensions of each plate 31, 32 and 33, and their corresponding slots 312,313, 322, 333, 332, 333, as used in the device represented in Figure 1.1.

[032] Figure 1.12 is an isolated two-dimensional representation of one of the blocks 51 that make up the module for evaluating the spatial resolution 50 of MRI images of different modalities. In it, the dimensions are represented

Petition 870170066292, of September 6, 2017, p. 15/67

12/35 and hole arrangements 512 used in the evaluation of this parameter, which may have a circular shape (Figure 1.12 (A)), square ((Figure 1.12 (B)), as well as any other configuration that is useful for evaluating the resolution The corresponding holes 511 for inserting screws that fix this block to one of the plates 31 of the cut thickness module 30 are also shown.

[033] Figure 1.13 is a two-dimensional representation of a second block 52 that makes up the module for the evaluation of spatial resolution 50 of MRI images of different modalities, similar to that represented in Figure 1.12. Also shown are the corresponding holes 521 for inserting screws that secure this block to one of the plates 32 of the cut thickness module 30.

[034] Figure 1.14 is a two-dimensional representation of a third block 53 that makes up the module for evaluating the spatial resolution of MRI images of different modalities, similar to that represented in Figures 1.11 and 1.12. Also shown are the corresponding holes 531 for inserting screws that secure this block to one of the plates 33 of the cutting thickness module 30.

[035] Figure 1.15 is a two-dimensional representation of the plates 41 and 42 that make up the module for the evaluation of diffusion tensor images and diffusion-weighted images for reconstruction of tract and fiber images of living organisms or objects 40. In 1.15 (A) and 1.15 (B) are represented, respectively, the plates that constitute the major side 41 and the minor side 42 of the module for evaluating DTI in one of its possible configurations, as well as the holes

Petition 870170066292, of September 6, 2017, p. 16/67

13/35 for fixing bundles of tracts and fibers.

[036] Figure 2 is a representation, in the form of MRI images, of specific modules for the evaluation of spatial resolution, slice thickness, geometric distortions and SNR (coronal and axial views);

[037] Figure 3 is a representation, in MRI images, of the module for the assessment of the section thickness (coronal view), and also a representation, in graphical form, of the profiles obtained for the estimation of this parameter;

[038] Figure 4 is a representation, in RM image (axial section) of the module for measuring slice thickness and for measuring SNR;

[039] Figure 5 is a representation, in coronal views, of the module for evaluating DTI and the module for evaluating geometric distortions;

[040] Figure 6 is a representation, in coronal MRI images, of the blocks that constitute the module for estimating spatial resolution;

[041] Figure 7 is a graphical representation of the mean values of MD, FA, RA and VR, as well as the corresponding values of coefficients of variation (CV) for these DTI indices;

[042] Figure 8 is a graphical representation of average values of CS, CL and CP and the corresponding CV values for these DTI indices;

[043] Figure 9 is a representation of reconstructed tracts from DTI acquisitions using the device in Figure 1.1;

[044] Figure 10 is a representation of other treatments reconstructed from DTI acquisitions using a coil.

Petition 870170066292, of September 6, 2017, p. 17/67

14/35 32-channel skull;

[045] Figure 11 is a representation of yet other reconstructed tracts from DTI acquisitions ranging TE and NSA, for 8 and 32 channel skull coils; and [046] Figure 15 is a representation of other reconstructed tracts from acquisitions using 8-channel skull coils, for bundles of fibers wrapped with heat-shrink tubes.

DETAILED DESCRIPTION OF THE INVENTION:

[047] The object of the present invention will now be described in more detail and explained based on the attached drawings, which are merely exemplary and non-limiting, since adaptations and modifications can be made without, therefore, escaping the scope of protection claimed.

[048] As can be seen from Figures 1.1 and 1.2, device 100 basically comprises at least one container of any geometric shape, represented in this case by a cylinder 101, into which modules 20, 30 are inserted , 40 and 50, which make it possible to evaluate different performance aspects of an MRI equipment. It should be noted that, in a preferred embodiment, cylinder 100 is made of acrylic, has a height between 160 mm and 200 mm, preferably 180 mm, external diameter between 150 mm and 190 mm, preferably 170 mm and internal diameter comprised between 140 mm and 180 mm, preferably 160 mm. The geometries and dimensions used in the manufacture of the cylinder are shown in detail in Figures 1.7, 1.8 and 1.9.

Petition 870170066292, of September 6, 2017, p. 18/67

15/35 [049] Specifically for the QC of DTI, or other techniques that allow the reconstruction of images of tracts and fibers of the organism or objects, the device 100 is used, a first module consisting of a block 40, hollow , containing two major sides 41 and two minor sides 42. Each of the two major sides 41 and the two minor sides 42 of block 40 comprise a plurality of holes 43 equally spaced apart (preferably), as can be seen in the Figures 1.1 and 1.6 and also in the photographic representations of Figures 2, 3 and 4. The main function of these holes 43, is to enable the fixation of bundles of fibers 44 to simulate tracts and fibers of different thicknesses and in different arrangements, as illustrated in the Figures 1.6, and in the photographic representations of Figures 3 and 4. In addition to fixing the beams 44, the holes 43 of the first module are also used for the evaluation of geometric distortions in the image, which and they can lead to the deterioration of the shape of their holes and that can be detected in a qualitative and quantitative way. In a preferred embodiment, the block 40 of the first module is rectangular, the two largest sides 41 have dimensions of 100 x 100 x 5 mm (50,000 mm ³ ) and twelve circular holes 43 of 10 mm in diameter, while the two smaller sides 42 have dimension of 60 x 60 x 5 mm (18,000 mm ³ ) and nine 43 circular holes of 10 mm in diameter each, as shown in Figure 1.15. The holes 43 of the first module 40 make it possible to fix bundles of fibers 44 of different materials to simulate tracts and fibers of different thicknesses and in different arrangements. The implementation of this module and the device as a whole are not limited to the dimensions and geometries described

Petition 870170066292, of September 6, 2017, p. 19/67

16/35 in this paragraph.

[050] In addition, the geometric distortion can also be evaluated through a second specific module 20, and as can be seen in Figures 1.1, 1.2, 1.3, 1.10 and also in the photographic representations of Figures 2 and 3, this module consists of a second module plate 20, provided with a plurality of holes 21 of the second module equally spaced apart. In a preferred embodiment the plate of the second module 20 is rectangular, 120 x 100 x 5 mm (60,000 mm ³ ) in size, made of acrylic, and the holes 21 are circular and 10 mm in diameter, as shown in Figure 1.10 (THE). In Figure 1.10 (B) another form of embodiment of this module is shown, using square holes of 10 mm on the side, and the embodiment of the device is not limited to the dimensions and geometries described in this paragraph.

[051] Regarding the assessment of the section thickness, it can be performed in three directions, axial, sagittal and coronal, through the use of a third module 30 composed of at least one first 31, a second 32 and a third 33 plates of the third module equipped with slots (312, 313, 322, 323, 332, 333), as seen in Figures 1.1, 1.4, 1.5, 1.11 and also in the photographic representations of Figures 2 and 3. At least two of these plates are fixed on the third, forming a U. In a preferred embodiment the first 31, the second 32 and the third 33 plates of the third module are square, have 100 x 100 x 8 mm (80,000 mm ³ ) in dimension, and the slots they are diagonal and are 1 mm (312, 322, 332) and 2 mm thick (313, 323, 333) and 5 mm deep. The plates 31, 32 and

Petition 870170066292, of September 6, 2017, p. 20/67

17/35 each contain 2 holes of 3.2 mm in diameter, the pairs of holes on each plate being labeled 351, 352 and 353. These holes are intended for the insertion of pairs of M3 screws labeled respectively 311, 321 and 331, preferably nylon, for fixing blocks 51, 52 and 53, which make up the spatial resolution module 50, on each of the plates. However, the implementation of the device is not limited to the dimensions, geometries and fixation options described in this paragraph.

[052] For the assessment of spatial resolution in the axial, coronal and sagittal planes, a fourth module was devised

50, containing at least three blocks, each block (51, 52 and 53) of the fourth module containing at least four rows of holes, as shown in Figures 1.4, 1.5, 1.12, 1.13 and 1.14, as well as in the photographic representations of Figures 2 and 3. Each row consists of four equally spaced holes (preferably) and of the same diameter. The sets of holes for each of the three blocks are represented as 512, 522 and 532, in reference to the blocks

51, 52 and 53, respectively. In a preferred embodiment, blocks 51, 52 and 53 of the fourth module are 20 χ 25 χ 5 mm (2,500 mm ³ ) in size, the hole in the first row being circular and 0.5 mm in diameter, the hole in the second row being circular and of diameter equal to 1 mm, the hole of the third row being circular and of diameter equal to 2 mm and the hole of the fourth row being circular and of diameter equal to 2.5 mm. In addition to the configuration of this module using circular holes, as shown in Figures 1.12 (A), 1.13 (A) and 1.14 (A), square holes can also be used, their sides being identical to the diameters

Petition 870170066292, of September 6, 2017, p. 21/67

18/35 of the circular holes used, this configuration being represented in Figures 1.12 (B), 1.13 (B) and 1.14 (B). Each of the blocks also has pairs of holes 511, 512 and 513 for fixing each of them on one of the plates of the cut thickness module 30, in the configuration described in the previous paragraph. In a preferred configuration, fixation is done with M3 nylon screws. However, the implementation of this module, as well as the device, is not limited to the dimensions and geometries described in this paragraph.

[053] For preferred embodiment of device 100, making cylinder 101 and modules 20, 30, 40 and 50 properly positioned, modules 20, 30, 40 and 50 5 are assembled together on at least two spindles

119, these being fixed to the cover 110 of the device 100. These spindles 119, illustrated among others in Figure 1.1, allow each of the modules 20, 30, 40 and 50 to be positioned at the desired height inside the cylinder 101, through a plurality of positioning devices

120. Modules 20, 30, 40 and 50 can be used together or separately, according to the performance characteristics of the scanner to be evaluated. The SNR measurement is made from the image of the cylinder portion 101 that contains only the liquid that fills it, without the presence of modules 20, 30, 40 or 50 described above. Spindles 120 are preferably made of polyacetal and are 170 mm long and 6 mm in diameter. However, this is not the only possible configuration for fixing and positioning the modules inside the device.

[054] Device 100 for assessing DTI and other fiber reconstruction techniques, assembled with five bundles of fibers and a reference, all arranged inside the

Petition 870170066292, of September 6, 2017, p. 22/67

19/35 cylinder 101, which is filled with distilled water. Table 1.1 presents the description of the characteristics of the fiber bundles used in the tests of the device 100. Table 1.1: Characteristics of the fiber bundles used in the tests of the device 100.

Beam Code Number Diameter Diameter Length Fibers in of Beam Beam Fibers Fibers (mm) (mm) (mm) REF Heat shrink tube 5 mm long and 10 mm

mm diameter used as a reference

SPECTRA B1 positional 400 0.178 8 100 DYNEEMA 1 B2 300 0.40 10 100 DYNEEMA 2 B3 300 0.35 10 100 DYNEEMA 3 B4 300 0.40 7 60 KEVLAR B5 100 0.34 7 60

[055] To simulate tracts and fibers of the human body, bundles of fibers were constructed using commercially available multifilament fishing lines, such as Dyneema®, Spectra® and Kevlar®. For the present device 100, we opted for the use of fishing lines given their greater availability on the market in relation to the raw materials, which originate them. Dyneema® and Spectra® are trade names of ultra high molecular weight polyethylene, or UHMWPE (acronym for ultra high molecular weight polyethylene), commonly used for the manufacture of high-strength objects. According to the literature, UHMWPE fibers are useful for the construction of bundles that simulate tracts and fibers in the axons of neurons. Because they are fibers composed of multiple filaments, they prove to be useful, have anisotropic diffusion, following a preferential direction, analogous to what occurs in tracts and

Petition 870170066292, of September 6, 2017, p. 23/67

20/35 fibers of the organism. The parameters calculated from the images of these bundles are within the range of values observed for axonal tracts, unlike, for example, bundles constructed using linen, polyamide and silk. Kevlar® is a highly resistant polymer, polyamide fiber, also used in the manufacture of personal protective equipment, tires, military equipment and high-resistance fishing lines. This material proved to be useful for the construction of catheters for use in interventional procedures guided by MRI, without causing distortions or artifacts in the images.

Tests performed [056] The evaluation of the device 100 for QC in MRI was performed using an elaborated acquisition protocol that includes T1, T2, proton density (PD) and diffusion tensor images. The MRI scanner used was the Achieva® 3T, produced by Philips. The acquisition parameters are listed in Table 2.1. The total acquisition time for the protocol was, on average, equal to 20 minutes. Table 2.1: Acquisition parameters used in device tests.

Parameters Images Acquired from

Acquisition

T1 T2 DTI 1 DTI 2 Sequence of SE SE SE-EPI SE-EPI Wrists Field of 180x180x 180x180x 180x180x 180x180 Vision (FOV) 200 200 200 200 (mm ³ ) Dimension of 2 x 2 x 2 2 x 2 x 2 2 x 2 x 2 2 x 2 x 2

Voxel (mm ³ )

Petition 870170066292, of September 6, 2017, p. 24/67

21/35

Thickness of Slice (mm) 2 2 2 2 Number of slices 100 100 100 100 TR (ms) 400 - 700 65 7000 7000 TE (ms) 65 400 65 65 b (s / mm ² ) 1000 2500

[057] The modules 20, 30, 40 and 50 of the device 100 were positioned at a distance of at least 10 mm from each other, so that they could be adjusted as needed. The first module 40 for assessing DTI, contained at least three bundles of parallel fibers, two from Dyneema® fishing lines and one from Spectra® fishing lines.

[058] The slice thickness was measured using the

T1 images of device 100, taking the profile of rips diagonal 312, 313, 322, 323, 332 and 333 of the three plates square 31, 32 and 33 of third module destined to this end. Given away that the slices this Image had 2 mm of

thickness, only gaps with this dimension (313, 323 and 333) were used for this evaluation. The slice thickness is defined by the width at half height (FWHM), of the profile obtained in a region of interest (ROI), which includes the corresponding slot of the square plate.

[059] SNR can be measured by taking the average value of the pixel intensity in an ROI and the noise as the standard deviation, σ, of the pixel intensity of one or more ROIs in the air. The SNR is then calculated using the following expression:

ç

SNR = 0.655 -, (3.3) σ

Petition 870170066292, of September 6, 2017, p. 25/67

22/35 [060] The factor 0.655 comes from the Rician distribution of background noise in a magnitude image, tending to a Rayleigh distribution as the SNR tends to zero. This is because the noise variations, which can be negative and positive, are all taken as positive, which artificially reduces σ. In this study, SNR was assessed using Equation 3.3, taking a slice of the acquired axial T1 image. This slice shows only the water that fills the cylinder, not containing solid internal structures of the device 100. Slices containing only the phantom filling liquid can also be used in the QC of each of the acquired DWI.

[061] Geometric distortions due, for example, to possible inhomogeneities of B0 or instability of the gradients, were qualitatively evaluated with the images of the hole pattern plate 20 and the hole pattern 43 present in block 40 where the fiber bundles 44 were fixed. The geometry of the holes (21 and 43) was analyzed in the T1 and T2 images.

[062] The spatial resolution was estimated from blocks 51, 52 and 53, with different hole sizes (512, 522 and 532), located above the square plates used to measure the slice thickness. We evaluated the smallest hole size displayed in each block to define the spatial resolution of the image.

[063] The QC of DTI acquisitions, for the evaluation of residual error and detection of significant outliers (signal distortions), was performed using the ExploreDTI software. DTI index values were calculated for each fiber bundle, for acquisitions using b = 1000 s / mm ² and b = 2500

Petition 870170066292, of September 6, 2017, p. 26/67

23/35 s / mm ² . The values of coefficients of variation (CV) of these indexes were then estimated to assess the stability of the device by varying b and were reconstructed.

Test results [064] Figure 2 shows the images and then the treatments

T1 of the developed device where it is possible to check the performance parameters of the scanner, through specific modules for the evaluation of the spatial resolution, slice thickness, geometric distortions and

SNR, which can be estimated using the homogeneous portion of the device. Still in relation to Figure 2, it is possible to specifically verify:

Figure 2 (a):

the ramps to calculate the slice thickness;

Figure 2 (b):

the grid pattern for evaluating geometric distortions;

Figure 2 (c): the bottom to determine SNR;

Figures 2 (d) and 2 (e): the blocks to evaluate the spatial resolution and also the bundles of fibers (view

A section of a three-dimensional rendering of the device is shown;

Figures

2 (g), 2 (h) and 2 (i):

the bundles of fibers longitudinally;

Figure 2 (j) the block where the fibers are fixed, which also allows the detection of geometric distortion patterns.

065] Additionally, and as it is possible to verify in the

Figure 2, the image

T1 of the device does not present significant artifacts that could compromise the identification of the structures of each module. Although there are some regions of

Petition 870170066292, of September 6, 2017, p. 27/67

24/35 non-homogeneities in the background, these can be minimized using compensation methods and even solutions based on Copper Sulfate II (CuSO4), which improve the SNR. However, it should be noted that, before any kind of quantification in the present test, a visual inspection of the images was made.

[066] In all tests, it was decided to fill the device with distilled water, given that the diffusion of water molecules is the object of study in DWI and DTI. However, conventional MRI phantom filling solutions, such as CuSO4, can improve the SNR and consequently the characterization of other MRI parameters evaluated with the phantom. The acceptability of the measured parameters was based on standards established for other phantoms, such as the phantom of the American College of

Radiology (ACR) and the American Association of Physicists in

Medicine (AAPM), although the structures present in the device described here differ from them in size and shape.

[067] Figure 3 shows the T1 image of the third module for assessing the section thickness and the profiles obtained to estimate this parameter. Through Figure 3 (a) it is possible to identify the ROIs in the areas delimited in yellow. The graphs illustrated in Figures 6 (b) and 6 (c) identify the profiles found for the ROIs at the top and bottom of the image, for the 2 mm diagonal gap.

[068] The slice thickness was measured from the FWHM of the 2 mm crack profiles. The mean and standard deviation of FWHM in both ROIs were taken. From the ROIs in the image in Figure 2, we obtained a slice thickness of 2.005 ± 0.1768 mm, differing approximately 0.5% from the real value

Petition 870170066292, of September 6, 2017, p. 28/67

25/35 the thickness of the slice used to acquire the images, 2 mm. According to the literature, specifically the “American College of Radiology. ACR 2015 ”and“ American Association of Physicists in Medicine. AAPM 2015 ”, differences of up to 0.7 mm are acceptable, so the result presented is within the established standards.

[069] Figure 4 shows axial T1 images of the device, used to estimate other performance parameters of the MRI scanner, going beyond the DTI QC. Basically, it is possible to see laterally the ramps for measuring the slice thickness and the polyacetal spindles 119, which did not generate any kind of image artifacts. SNR was measured using the ROIs shown in the center of the device and outside the device, as shown in Figures 7 (a) and 7 (b), respectively.

[070] From the image and the ROI shown in Figure 4, an SNR value of 217 was obtained, for a signal of 1350 average intensity in the center ROI and a background standard deviation of 4, which is acceptable according to with SNR references found in Dietrich O, Raya JG, Reeder SB, Reiser MF, Schoenberg SO. Measurement of signal-to-noise ratios in MR images: influence of multichannel coils, parallel imaging, and reconstruction filters. J Magn Reson Imaging. 2007; 262: 375-85, according to which this parameter must be in the range 130-27.

[071] This image slice could also be used to estimate the integral uniformity of the image, which should be calculated in future tests. If the homogeneity of the image is not adequate, the SNR can be estimated more precisely using the method described by

Petition 870170066292, of September 6, 2017, p. 29/67

26/35 NEMA (National Electrical Manufacturers Association), which recommends the acquisition of two consecutive images for the calculation of this parameter by subtracting them.

[072] Figures 8 (a) and 8 (b) show the coronal views of block 40 that makes up the first module for evaluating DTI, whose hole pattern can also be useful for detecting geometric distortions. Figure 5 (c) shows the image of the geometric distortion module, a plate with a specific hole pattern for assessing distortions 20.

[073] It was found that both the plate with hole pattern 20 and block 40 where the bundles of fibers 44 are attached contribute to evaluate this image parameter. The image of the hole patterns showed a smooth intensity difference in one of its edges due to the position of its structures in the skull coil, structures that were close to the edge of the coil, which can be improved by changing its position in the spindle. Spindles 119 also allow the use of complete configuration or just one or more modules for the QC. Even when only DTI data are acquired, block 40 for fiber bundles 44 is useful for detecting geometric distortions due to parasitic currents, for example, which can degrade the results obtained in the quantifications.

[074] In relation to Figure 6, it has coronal sections of the device, in which it is possible to identify the small blocks (51, 52 and 53) of the fourth module to assess spatial resolution, according to the smallest hole pattern visually distinguishable. On both slices, two of the three blocks are displayed. The third block is seen in sagittal slices. A specific module to verify performance

Petition 870170066292, of September 6, 2017, p. 30/67

27/35 of the protocols for the detection of low contrast structures may also be implemented in the future.

[075] The efficiency of the DTI module for the QC of this imaging technique was evaluated by analyzing the DTI indices obtained, for the 8-channel coil, changing the value of the diffusion sensitization factor b. Figures 10 and 11 show the graphs of mean values and CV for each DTI index estimated in this study, for three of the fiber bundles described in Table 2.1.

[076] Additionally, the graphs in Figure 7 show that beam B2 has higher FA values. CV values for all DTI indices were low for all beams, indicating the long-term stability of the structures, as well as their applicability for cross-calibration studies, protocol design and improvement and also for routine QC in DTI. The graphs in Figure 7 illustrate the DTI indices and respective CV values for DTI acquisitions with b = 1000 s / mm ² and b = 2500 s / mm ^2, using the device, with Figures 10 (a) to 10 (d) show the average values found for MD, FA, RA and VR and Figures 10 (e) to 10 (h) show the CV values for these DTI indexes. The graphs in Figure 8 illustrate the DTI indices and respective CV values for the acquisition of DTI images with b = 1000 s / mm ² and b = 2500 s / mm ² , using the device, and Figures 11 (a ) to 11 (c) show the average values found for CS, CL, CP and Figures 11 (d) to 11 (f) show CV values for these DTI indexes.

[077] The Spectra® fishing line bundle had the lowest FA values, also having lower RA and greater MD

Petition 870170066292, of September 6, 2017, p. 31/67

28/35 than the others. Despite the fact that Dyneema® fibers have the best FA and SNR values for DTI QC, simulating axonal tracts and fibers, Spectra® lines were also tested for being composed of fibers of the same nature (UHMWPE). Spectra® lines proved to be useful for the construction of structures that simulate cardiac muscle fibers, with the values of MD and those of these muscle fibers.

[078] Figure 9 presents the

FA found next reconstructed fiber tracts from acquisitions of

DTI using the device 100, and for Figures 12 (a) to 12 (d) acquisitions with

1000 s / mm ² and for

Figures 12 (e) to

12 (h) acquisitions with

2500 s / mm ² . It should also be noted that the treatments illustrated in

Figures 12 (a) to 12 (e) were registered with images

T1, the treatments illustrated in Figures 12 (b) and 12 (f) were co-registered with the T2 images, the treatments illustrated in

Figures 12 (c) and 12 (g) were registered with DWI and the treatments illustrated in

Figures 12 (d) and

12 (h) were registered with the ADC map.

079] Additionally, the fact that device 100 has many other modules does not interfere with DTI and its indexes.

Thus, the results found in the present show that the device is useful for multiple modalities of MRI images. The assay / test evaluation of the same logic explored here can be applied to the development of phantoms that aim to evaluate, for example, the parameters verified here and also spectroscopy, functional or angiographic parameters, as well as positron emission tomography (PET) parameters, PET / RM scanners. Modules

Petition 870170066292, of September 6, 2017, p. 32/67

29/35 and or specific compounds for the evaluation of DWI can also be added to the device, as well as modules for evaluating the CNR and the ability of the MRI scanner to identify structures in low contrast conditions.

[080] Other reconstructions obtained with bundles of fibers composed of Dyneema® fishing lines are shown in Figures 13 and 14, where a 32-channel skull coil was used. It should be noted that the images illustrated in Figures 13 (a) to 13 (e) show reconstructed tracts registered with the T2 images and the images illustrated in Figures 13 (f) to 13 (j) show reconstructed tracts registered with ADC maps.

[081] Figure 11 shows reconstructed treatments for D | TI acquisitions, varying TE and the number of signal averages (NSA), for skull coils of 8 and 32 channels. Treatments are registered with T2, DWI images, ADC map and directional FA maps of the device, as well as illustrated separately. The images corresponding to Figures 14 (a) to 14 (j) refer to the DTI acquisitions varying TE and NSA, respectively, for the 8-channel skull coil and the images corresponding to Figures 14 (k) to 14 (t ) refer to reconstructions equivalent to the 32-channel skull coil.

[082] The reconstruction of treatments for acquisitions using the 8-channel skull coil, for bundles of fibers wrapped with heat-shrink tubes, can be seen in Figure 15, with the images shown in Figures 15 (a), 15 (b ), 15 (c), 15 (d) and 15 (e) correspond to the images registered with T1, DWI, ADC, PD map and directional FA color maps respectively. The Figure

Petition 870170066292, of September 6, 2017, p. 33/67

30/35

15 (f) corresponds only to the image of the reconstructed tracts.

[083] Tests with the DTI device object of the present invention showed that it is possible to measure DTI indices and other MR parameters using with a single device. In an image acquisition, with a total time of approximately 20 minutes. In the tests performed, it was possible to obtain information about the slice thickness, spatial resolution, geometric distortion, SNR and also about the stability of the calculated DTI indexes. The SNR value found was compatible with the application of the scanner in the clinical routine. SNR can also be measured only with the acrylic cylinder, without internal structures, filled with distilled water or even solutions based on CuSO4. In this case, it is possible to calculate this parameter for all slices in the image. The thickness of the slices and the spatial resolution were compatible with the thickness of the slice and the size of the voxel of the image protocol. Acquisitions in all directions can be made to check these parameters for all cutting orientations (axial, sagittal and coronal). Regarding geometric distortions, both the hole pattern plate and the hole pattern of the DTI block can be useful to detect this problem.

[084] Therefore, the developed device 100 proved to be useful for the low cost and relatively fast QC of many parameters of the MRI scanner. The fiber bundles used showed FA values close to those found for the main brain tracts, which did not show high CV when the b value was changed. Of all the tested fishing lines, Dyneema® is the most suitable for the construction of DTI phantoms, in agreement

Petition 870170066292, of September 6, 2017, p. 34/67

31/35 with results from previous studies using only Dyneema® fibers (raw material). For these lines, the DTI indices are stable over time, for different lengths or orientations of fiber bundles. In addition, this material has little memory when compared to others and, consequently, does not introduce artifacts in DTI.

[085] It is concluded, therefore, that the device 100 for the quality control of images of the diffusion tensor, of diffusion-weighted images for reconstruction of images of tracts and fibers and of other parameters in magnetic resonance of the present invention solves the technical problems to which it is addressed. In this context, a device and consequently a QC protocol is developed to assess the quality of the data obtained in acquisition of diffusion-weighted magnetic resonance images, DWI, and in images of the diffusion tensor, DTI, which additionally allows the evaluation of other quality parameters in magnetic resonance, such as: section thickness, signal-to-noise ratio, spatial resolution and detection of geometric distortions, as well as the QC in DWI acquisitions that aim to reconstruct tracts and fibers of human biological tissues, animals, vegetables and even objects, promoting, therefore, more reliable results with more practical procedures than those achieved with the use of tools known in the current state of the art.

[086] Those skilled in the art will value the knowledge presented here and may reproduce the invention in the modalities presented and in other variants, covered by the scope of the attached claims.

Petition 870170066292, of September 6, 2017, p. 35/67

32/35

LEGEND OF THE DESCRIPTIVE REPORT AND LIST OF ELEMENTS OF THE

FIGURES

100 - Device

101 - Main cylinder

110 - Upper cylinder base (cover)

111 - Cover fixing screws on the cylinder body

112 - Sealing screws for the cylinder access holes

113 - Cylinder body

114 - Base for fixing the spindle

115 - O'ring sealing ring

116 - Base for fixing the cylinder cover

117 - Side surface of the cylinder

118 - Lower cylinder base

119 - Spindle for fixing and positioning the modules inside the cylinder

120 - Thread for positioning and fixing modules inside the cylinder

121 - O'ring fitting cavity

122 - Fixing screw holes for access seals on the cylinder cover

123 - Holes for fixing screws for fixing the cylinder cover to the body

- Module for evaluation of geometric distortions

- Holes for evaluation of geometric distortions

- Module for evaluation of cutting thickness

- First cutting thickness module plate

311 - Block fixing screws for spatial resolution evaluation on the plate 31

312 - Slit 1 mm thick in plate 31 for

Petition 870170066292, of September 6, 2017, p. 36/67

33/35 cutting thickness assessment

313 - Slit 2 mm thick in the plate 31 to assess the thickness of the cut

351 - Holes for fixing screws for positioning one of the spatial resolution blocks on plate 31 of the cutting thickness module

- Second cutting thickness module plate

321 - Block fixing screws for evaluating spatial resolution on the plate 32

322 - Slit 1 mm in thickness on the plate 32 for evaluation of thickness in cut 323 - Slit 2 mm in thickness on the plate 32 for evaluation of thickness in cut

352 - Holes for fixing screws for positioning one of the spatial resolution blocks on plate 32 of the cutting thickness module

- Third cutting thickness module plate

331 - Block fixing screws for the evaluation of spatial resolution on the plate 33

332 - Slit 1 mm in thickness on the plate 33 for evaluation of thickness in cut 333 - Slit 2 mm in thickness on the plate 33 for evaluation of thickness in cut

353 - Holes for fixing screws for positioning one of the spatial resolution blocks on plate 33 of the cutting thickness module

- DTI module (main block)

- Larger side of the block that makes up the DTI module

- Smaller side of the block that makes up the DTI module

- Holes for fixing devices for evaluation

Petition 870170066292, of September 6, 2017, p. 37/67

34/35 of the DWI and DTI

- Fiber bundles for DTI assessment

- Module for the evaluation of spatial resolution (there is no physical structure directly joining the blocks that compose it)

- Block 1 of the spatial resolution module

511 - Holes for fitting the screws for fixing the spatial resolution block on the cutting thickness plate 31

512 - Holes present in block 51 to assess spatial resolution (holes of all dimensions used)

- Block 2 of the spatial resolution module

521 - Holes for fitting the screws for fixing the spatial resolution block on the cutting thickness plate 32

522 - Holes present in block 52 to assess spatial resolution (holes of all dimensions used)

- Block 3 of the spatial resolution module

531 - Holes for fitting the screws for fixing the spatial resolution block on the cutting thickness plate 33

512 - Holes present in block 53 to assess spatial resolution (holes of all dimensions used)

AAPM - American Association of Physicists of Radiology ACR - American College of Radiology

ADC - Apparent Diffusion Coefficient

Stroke - Stroke b - diffusion sensitization factor

CL - Linearity Coefficient

Petition 870170066292, of September 6, 2017, p. 38/67

35/35

CNR - Contrast to noise ratio or Contrast to noise ratio

CP - Coefficient in planicity CS - Coefficient in Sphericity CV - Coefficient in variation

DT / DTI - Diffusion or Diffusion Tensioner images

Imaging

DW / DWI - Difusion-Weighted Imaging Weighted Magnetic Resonance Images

FA - Fractional anisotropy

FWHM - Width at half height

MD - Average tensor diffusivity

NSA - Number of signal averages

PD - Proton density

PET - Positron Emission Tomography

RA - Relative anisotropy

RF - Radio Frequency

MRI - Magnetic resonance imaging

ROI - Region of interest

SNR - Signal-to-noise ratio or Signal-to-noise ratio

TE - Time to echo

UHMWPE - Ultra-high molecular weight polyethylene or

Ultra High Molecular Weight Polyethylene

VR - Volume ratio σ - Standard deviation

Claims (3)

1. Device for quality control of diffusion tensor images, diffusion-weighted images for reconstruction of tract and fiber images and other parameters in magnetic resonance FEATURED by the fact that it comprises: - at least one cylinder (101); - a first module (40) arranged on the cylinder (101); - a second module (20) disposed on the cylinder (101); - a third module (30) disposed in the cylinder (101) and cooperating with the first (40) and the second (20) modules; - at least two spindles (119) for cooperation of the first (40), the second (20) and the third (30) modules with the cylinder (101); - a fourth module (50) arranged on the cylinder (101); and said device (100) further comprising a cap (110) cooperating with the cylinder (101) for fixing the spindles (119) and a liquid solution to allow the diffusion of water molecules. 2. Device according to claim 1, CHARACTERIZED by the fact that the cylinder (101) has a height understood in between 160 mm and 200 mm, diameter external understood in between 150 mm and 190 mm, and diameter internal understood in between 140 mm and 180mm. 3. Device in wake up with claim 2, CHARACTERIZED BY fact cylinder (101) have 180 mm in height, external diameter 170 mm and internal diameter of 160 mm. 4. Device in wake up with claim 1, CHARACTERIZED BY fact that the first module ( 40) Petition 870170066292, of September 6, 2017, p. 65/67 2/3 consists of a block (40) with two smaller sides (41) and two larger sides (42), with the larger sides (41) and the smaller sides (42) comprising a plurality of holes (43) equally spaced, so that the holes (43) allow the fixation of fiber bundles (44). 5. Device in agreement c om The claim 4, CHARACTERIZED BY THE FACT that the two larger sides ( 41) have dimension of 100 χ 100 χ 5 mm and twelve holes 43) 10 mm circular in diameter, and the two smaller sides (22) have a dimension of 60 χ 60 χ 5 mm and nine circular holes (43) with a diameter of 10 mm. 6. Device according to claim 1, CHARACTERIZED by the fact that the second module (20) has a plurality of holes (21) of the second module equally spaced apart. 7. Device according to claim 6, CHARACTERIZED by the fact that the second module (20) is rectangular and has a dimension of 120 x 100 x 5 mm and the holes (21) of the second module (20) are 10 mm circular in diameter. 8. Device according to claim 1, CHARACTERIZED by the fact that the third module (30) has at least one first (31), a second (32) and a third (33) plates of the third module (30) equipped with slits (312), (313), (322), (323), (332), (333) and cooperating with each other in order to form a U. 9. Device according to claim 8, CHARACTERIZED by the fact that the first (31), the second (32) and the third (33) plates of the third module (30) are square, dimension 100 χ 100 χ 8 mm and the cracks (312 and Petition 870170066292, of September 6, 2017, p. 66/67 3/3 313), (322 and 323), (332 and 333) are diagonal to each other, two by two as mentioned, and are 1 mm and 2 mm thick and 5 mm deep. 10. Device according to claim 1, CHARACTERIZED by the fact that the fourth module (50) has at least three blocks (51) (52) and (53), each block with at least four rows containing each of them four holes equally spaced and of the same diameter, the holes in each block being labeled (512), (522) and (532). 11. Device according to claim 1, CHARACTERIZED by the fact that the spindles (119) allow the modules (20, 30, 40 and 50) to be positioned at the desired height inside the cylinder (101), through a plurality positioning devices (120). 12. Device according to claim 1, CHARACTERIZED by the fact that it contains two or more modules that allow to evaluate the QC of DWI and DTI images together with other performance parameters of the MRI scanner.