Test phantom to control image quality in high local resolution NMR imaging

The invention relates to test structures (phantoms) for checking and evaluating the imaging properties of an imaging device, the imaging principle of which is based on magnetic resonance, consisting essentially of a whole set or individual periodically arranged structural elements of different sizes, which are located in a container that is equipped with an NMR- active material such as copper sulfate or manganese chloride solution is filled.

NMR microscopy⁵⁶ allows non-invasive, non-destructive, quasi-continuous imaging with a relatively high spatial resolution (up to approx. 8 x 8 x 50 µm³, in some cases also underneath®) on living structures and materials. The spatial resolution is also determined by the properties of the measurement object (eg line width and state of motion) in addition to the apparatus conditions.

In order to be able to compare and evaluate NMR imaging devices with regard to their imaging performance, in particular the achievable spatial resolution, phantoms defined with regard to their dimensions are required which contain structures in the resolution range of interest. Image test structures are also an indispensable prerequisite as a quality criterion for system improvements aimed at finer spatial resolutions. They also serve in the sense of continuous quality control to maintain an image quality that has already been achieved.

So far, phantoms have been described for various quality parameters of NMR tomographic images, for example for sensitivity and linearity^7,10. Several patents describe the control and quantification of layer thickness, position, inclination and deflection¹⁰·¹²·¹³·¹⁴. Ej_nGe publications relate to various NMR active materials such as deuterium oxide⁷, Hydrogels⁸, Polymer gels¹⁸ or polyvinyl alcohols (liquid and frozen)⁹. There are also descriptions for phantoms for recording the spatial course of the magnetic field¹⁵. Few works open up the possibility of recording several important, different quality parameters at the same time, eg: signal and contrast-noise ratio, geometric distortion, slice parameters, T1, T2, spin density, stability etc. and spatial resolution^8,10.

Up to now, resolution phantoms have mainly been used for clinical MR tomographs. They essentially consist of cylinder-like structures or holes and grooves with a defined gap spacing and opening width of at least approx. 0.6 mm¹¹. The structures were considered resolved when clear differences were observed in the intensity maxima and minima in the MR image of the grating. Test plates oriented orthogonally to one another allow the simultaneous recording of certain NMR quality parameters in two independent spatial directions¹⁷. Improvements to these phantoms consisted of test structures of various sizes⁹, also a whole set of grids of different periodicity. These allow the intensity modulation between the lattice gaps to be recorded as a function of the structure width^1,2,3. Furthermore, test structures are described which are based on the introduction of rods or plates into a cavity with a housing made of plastic.¹⁹

What these descriptions have in common, insofar as they are explicitly mentioned, is a typical smallest structure size of down to approx. 1 mm for the openings.

Substantially different from the previous publications, the subject matter of the invention has cavities in a base material on a semiconductor or ceramic base with very small structure widths of approx. 1-600 μm in a multiple periodic arrangement, for example, positioned perpendicular to one another. The very small structure widths can only be achieved through special manufacturing processes. In a special embodiment, structural elements are described which are characterized by particularly high structural depths in relation to the lateral (in the plane of the selected NMR layer) dimensions of the structural elements (high aspect ratio).

This is achieved in that the structural elements consist of ceramic or semiconductor or lacquer or polymer base material, from which cavities with structure sizes smaller than 600 each located in a single block of the base material.

The object of the invention consists essentially of a whole set or individual grids, consisting of periodically arranged columns or openings in a solid base material made of semiconductor or ceramic material, e.g. Silicon or gallium arsenide, or lacquers or polymers .. The gap distance can typically vary between approx. 1 μm and approx. 600 μm. The periodic openings, the spaces between the solid base material, are filled with an NMR-active liquid, e.g. Water solutions filled. In NMR imaging, for example in the form of a profile in one direction, the phantom ran periodic brightness distributions that allow the recording of the modulation, the intensity maxima and minima of the image intensity.

The following expression is understood as modulation M of an intensity distribution:

M = (Imax -lmin) / (lmax + Imin) (1)

AT 406 010 Β

Imax denotes the maximum of the intensity, as can be observed in an intensity profile (picture) over the lattice structure in the area of ​​the stomata due to the NMR-active liquid present there. Imine is the minimum observed in the profile in the area of ​​the base material that should not be observable by NMR. Is the object specifically made up of periodic structures, e.g. the spatial frequency K = 1 / a (a denotes the period of the structure), then the modulation transfer function M (K) denotes the modulation M as a function of the spatial frequency K. It allows a quantitative evaluation of the spatial resolution of the system, e.g. by defining a threshold value M._crit for modulation, e.g. M_cr"= 50%. The associated spatial frequency provides the spatial resolution 8 ^ ,, / 2.

As a single example for the implementation of the test structures / phantoms listed above, a set of grids is described which comprises 9 different gap spacings iB between 4 μm and 1024 μm (see Fig. 1). Two sets of grids (2 and 3) which are identical with regard to the gap spacing are arranged perpendicular to one another in order to be able to detect the modulations in two independent spatial directions in the NMR image at the same time (e.g. the frequency and phase coding direction for spin warp, 2DFT or back projection methods).

Also shown are several wedge-shaped gap openings (4) arranged concentrically around a point of symmetry in the base material (1). During the measurement, the actual test structure is located in a container which is filled with an NMR-active liquid, eg a copper sulfate-water solution. Copper sulfate leads to a reduction in T1 times. This shortens the measurement times.

In the following (overview of the figures), a possible configuration of the test structures described according to Figures 1 and 2 is described in more detail;

1 example of a test phantom for detecting the spatial resolution in NMR microscopy: top view

FIG. 1 shows two structures that are fundamentally different for the qualitative and quantitative evaluation of the spatial resolution achieved by the imaging system. Firstly, two sets of cavity structures (grid systems 2 and 3) are arranged perpendicular to one another in a base material (1) in order to be able to detect two independent image coding directions at the same time. Each grid system consists of a set of grids, which in turn consist of several columns (cavities) that accommodate the NMR-active substance. The gap width and spacing vary in the example between a width of 1024 µm and 4 µm for different grids. The respective gap sizes are entered above or next to the grid. Second, a test structure (part 4) with concentrically arranged wedge gaps can be integrated together with the grid structures in a phantom. Periodically with the angle of rotation, the solid base material and the NMR-active substance alternate in the circular segment (= wedge) -shaped cavities. A circular profile in the NMR image leads to a periodic modulation of the intensity with decreasing periods with smaller radii.

In Fig. 2, a longitudinal profile through the test structure described in Fig. 1 can be seen between the grating with a structure width of 256 μm, the intervening columns and the grating with a period of 32 μm. In this special embodiment, the necessary edge steepnesses become clear, which are associated with a high structure-depth-to-width ratio (aspect ratio), particularly in the case of the smaller lateral structure widths less than 64 μm. A possible choice of the layer structure within the framework of the NMR microscopic imaging of the test structure is shown in dashed lines.

The test structures can be produced by one or more of the methods described below: e.g.

1. ) The gaps can be made from the mechanical fine cutting, milling, sawing or cutting processes

Base material to be cut.

2. ) The gap structures are created with the help of physical ablation processes (e.g. laser or electron beam cutting).

3rd ) With the help of chemical (e.g. chemical etching, especially anisotropic etching)⁴) or physical (eg plasma etching, RIE process) processing using lithographic processes using masks., the base material is structured.

4th ) With the help of optical, X-ray, ion or electron beam lithographic processes on or in lacquer, plexiglass or polymer-like layers.

In contrast to previous inventions in the field of NMR imaging, the phantom is characterized by the extraordinarily short distances (fine structure) and cavity sizes of approx. 1-600 μm adapted for NMR microscopy. These structure sizes can only be produced with the help of special processes. In a special embodiment, large structure depths (high aspect ratio) of at least 150 μm are present at the same time in relation to the lateral spacings between the structure elements. These structure depths are necessary for a sufficiently high spin number in the detected volume as well as in the

AT 406 010 Β

Choosing a slice guide for MR imaging in the plane of the grating structures is very important. The test structures described can also be integrated as individual parts into an existing phantom, which has options for measuring other quality parameters, such as sensitivity, layer thickness, etc. It is also possible to produce a test phantom that does not have the individual test cavity elements on one piece of base material, but on several pieces.

Reference publications

Podo, E. et al.

Magn. Res. Imag. 6, 173-222 (1988)

Lerski, RA, Mc Robbie, DW

Eurospin II, Magnetic Resonance Quality Assesment Test Objects,

Instructions for use (1992)

Lerski, RA et al.

Magn. Res. Imag. 11, 809-840 (1993)

DL Kendall

Applied Physics Leader Vol. 26 No.4 1975

P. Mansfield, PG Morris NMR Imaging in Biomedicine in Advances in Magnetic Resonance ed. JS Waugh, Supplement 2,

Academic Press London (1982)

PT Callaghan Principles of Nuclear Magnetic Resonance Microscopy

Clarendon Press Oxford (1991)

<td> 7</td><td>US 4777442 A</td><td colspan="2">Rosenthal, MS</td>

<td> 8</td><td>US 4729892 A</td><td>Beall, PT</td><td></td>

<td> 9</td><td>US 5023185 A</td><td> 10.12.91</td><td>Nambu Nippon oi</td>

<td> 10</td><td>US 4888555 A</td><td> 19.12.89</td><td>Vaughan.JT</td>

<td> 11</td><td>US 462 5168 A</td><td> 25.11.86</td><td>Meyer, A.C.</td>

<td> 12</td><td>US 469 2704 A</td><td> 08.09.87</td><td>Gray, JE</td>

<td> 13</td><td>US 503 6280 A</td><td> 30.07.91</td><td>Chesavage, YES</td>

<td> 14</td><td>US 461 8826 A</td><td>Smith SL</td><td></td>

<td> 15</td><td>DE 40 044 184 C2</td><td colspan="2">Schmitt F.</td>

<td> 16</td><td>DE 195 11 124 A1</td><td>Gurvich,</td><td>Victor</td>

<td> 17</td><td colspan="3">USPTO No. 4644276 2/17/87 Sierocuk, TJ</td>

<td> 18</td><td>Gore, JC et. al.</td><td></td><td></td>

<td></td><td colspan="3">Med. Phys. 24 (9): 1405-1408 1997</td>

<td> 19</td><td>U.S. 4,613,819 A</td><td> 23.09.86</td><td>Chiu</td>

Claims