Phantoms, phantom and physical measurements in radionuclide scintigraphy

AstroNuclPhysics Nuclear Physics - Astrophysics - Cosmology - Philosophy Physics and nuclear medicine

PHANTOMS and phantom measurements in nuclear medicine
1. Introduction - the essence of phantoms, types of phantoms
2. Phantoms of organs
3. Phantoms for gamma camera testing
3.1. Testing and calibration of scintillation camera image homogeneity
3.2. Measurement of the positional resolution of the camera
3.3. Gamma Camera Imaging Linearity Analysis
3.4. Tomographic phantoms for SPECT, PET, CT
4. Dynamic phantoms
4.1. Modeling of periodic cardiac pulsation
4.2. Modeling of central hemodynamics
4.3. Phantom measurement of esophageal passage dynamics

1. Introduction - what are phantoms
Radiological imaging methods (X-ray diagnostics, nuclear medicine - scintigraphy, NMRI magnetic resonance imaging, ultrasonography) examine the structures and processes hidden inside the organism . The question therefore arises as to whether the relevant images capture these invisible structures precisely and objectively and are happening inside the body? For experimental testing of these aspects of imaging diagnostics, as well as for measuring physical parameters, imaging properties and for calibration of used devices, special precisely and reproducibly made aids, so-called phantoms, are used .
  The word phantom (Greek phantasm = vision, spirit, ghost) has a number of common meanings: delusion, ghost, vision, deceptive phenomenon - imaginary, fictional. Here we will use a somewhat special meaning of this word - an artificial fact , a model .
   The main task of radionuclide measurements and imaging in nuclear medicine is the analysis of structures and processes in the body through external detection of gamma radiation emanating from the body after the application of a suitable radio indicator - Chapter 4 " Radionuclide scintigraphy ". A number of side effects can be used to detect this g radiation and to evaluate the results, such as: statistical fluctuations, imperfect camera resolution, tissue absorption in the tissue, Compton scattering of g radiation, the influence of the tissue background (mostly inhomogeneous), the difference between the actual dynamics of the studied processes in the organism from the mathematical model, etc. These circumstances may adversely affect the accuracy of the measured parameters and the whole scintigraphic diagnostics. What are the details of the actual structures of the tissues and the actual course of the dynamics of events deep within the organism , a priori we do not know. The same is true for other imaging methods.
   Phantoms can play an important role in the exact approach to the development and verification of some radiological imaging methods (especially scintigraphic or X-ray). By phantom here we will understand an artificial modelof a given system, which simply expresses certain important anatomical features, the distribution of the radio indicator in the organism, or the dynamics of movement and time changes. Compared to the real organism, in which we investigate complex, variable and previously unknown anatomical conditions or distribution of the radioindicator, the phantom has two basic advantages for study purposes:

In phantom measurements, we can then compare these actual parameters and structures of the phantom with the images and with the values ??obtained by analysis of phantom radionuclide measurements. A necessary (unfortunately not sufficient) condition for the correctness and exactness of each measuring method is that it must work well on the relevant phantom ! Only then can other secondary effects in a given biological situation be considered. Otherwise, it would be just empiricism , with the possibility of erroneous results ....

Types of phantoms
Phantoms in nuclear medicine can basically be divided into three groups:
a) static phantoms of specific organs; b) phantoms for calibration and measurement of physical parameters of imaging devices (gamma camera, X-ray CT); c) dynamic phantoms .

2. Phantoms of organs
These phantoms mimic some typical anatomical shapes and distributions of a radioinducer in a given organ, in a physiological or pathological situation. The best known is the phantom of the thyroid gland , which is a vessel in the shape of a thyroid gland containing protrusions and depressions inside simulating cold and hot nodules. The vessel is filled with a solution of the appropriate radionuclide ( 99m Tc or 131 L) and imaged with a scintillation camera. We then assess how faithfully the individual details of the inhomogeneity of the radioindicator distribution in the phantom are captured in the scintigraphic image. At the same time, we try to achieve such conditions (camera settings, collimator, image matrix, accumulated number of pulses, image modulation) that the recognizability of even small lesions is as good as possible.

Phantom of the thyroid gland
Phantom of the thyroid gland
(thyroid Picker phantom)
Scintigraphic image of a
thyroid phantom filled with
99m Tc solution using a gamma camera
The same phantom displayed using a
motion scintigraph
Fig.2.2. Scintigraphic image of the thyroid phantom accumulated with different preselection of the number of pulses - with different number of gamma photons

Fig.2.3. Scintigraphic images of a thyroid phantom filled with 99m Tc ( top ) and 131 I ( bottom ), imaged using a Pinhole collimator , Low Energy HR and High Energy . A more detailed discussion is in 4.2, section " Scintigraphic collimators ".

Phantoms of other organs - liver, brain, heart are based on a similar principle ... An interesting option are tomographic phantoms in the form of anatomical inserts imitating brain or heart structures, which are inserted into tomographic test phantoms of the Jasczak type (see " Tomographic phantoms " below) in place of the original test rods and beads.

3. Phantoms for Gamma Camera Testing
Nuclear medicine, as a field of measurement and imaging, cannot do without control and calibration methods and aids. The category of phantoms in nuclear medicine also includes aids and radiation sources used for calibration and testing of imaging properties of scintillation cameras. Parameters quantifying the imaging properties of scintillation cameras are defined and discussed in Chapter 4 " Radionuclide scintigraphy ", 4.5 " Physical parameters of scintigraphy - quality control and phantom scintigraphic measurements " of the book " Nuclear physics and ionizing radiation physics ".
The simplest "phantom" a point source (emitter) of a suitable radionuclide (most often
99m Tc, or 57 Co, 131 I, 18 F), which is placed in the appropriate place of the camera's field of view; when placed at a sufficient distance from the gamma camera crystal without a collimator, it functions as a source of homogeneous radiation. They also use various other geometric structures - tubes, balls, cylinders filled with radionuclides. Important test phantoms are planar homogeneous sources - see below .

3.1. Testing and calibration of scintillation camera image homogeneity
The physical-technical parameter of camera image homogeneity is defined and discussed in more detail in 4.5 " Physical parameters of scintigraphy - image quality and phantom measurements " in the section " Homogeneity (uniformity) of camera field of view ". Two methods are used to test and calibrate the homogeneity of the gamma camera :
1. Internal homogeneity - point source
To test the internal ( intrinsic ) homogeneity of the camera's field of view and its calibration, remove the collimator and point source from the camera
(in practice it can be solution 99mTc with an activity of several MBq of small volume in a bottle or small syringe) is placed as far as possible (approx. 2 meters) in the middle under the camera *) - Fig.3.1.1 left. The camera crystal is thus irradiated practically homogeneously from this gamma radiation source , so that the resulting scintigraphic image should also be homogeneous. In the image, we can then assess (or quantitatively evaluate) any deviations from the homogeneous distribution. In case the homogeneity of the field of view is not satisfactory, in this arrangement we can calibrate the homogeneity of the scintillation camera display (tuning of photomultipliers and creation of a correction matrix), for which modern digital cameras have the appropriate software procedure.
*) The distance of the point source from the camera crystal - Geometric correction
Intensity I
g -radiation activity from a point source A at a distance r is given by the general relationship I (r) = G  . A / 4 p r 2  , where the coefficient G indicates the number of gamma quanta emitted by the radionuclide during one radioactive transformation. Therefore, if we place the point source at a distance h from the center of the scintillation crystal of the gamma camera (which we choose as the beginning of the planar coordinate system [x, y]), the crystal will be irradiated with intensity I (x, y) at each of its coordinates coordinates (x, y). = G  . A / [4 p ( h2 + x 2 + y 2 )]  (follows from the trigonometric analysis of the distances between the emitter and the x, y sites of the crystal) . The places of the crystal farther from the center will therefore be irradiated a little weaker than the central part. If the distance h is many times greater than the dimensions of the crystal, this effect is practically negligible - the crystal is irradiated almost homogeneously (as mentioned above) . However, at smaller distances h of the point source, for accurate measurement and calibration of homogeneity it is necessary to make a correction in the obtained image by the coefficients K (x, y) = ( h 2 + x 2 + y2 ) / h 2 . This is used with some SPECT cameras when calibrating homogeneity using a point source, inserted between both detectors only about 80 cm apart.
Note: Using a point source, we can also measure the overall resolution of a camera with a collimator: we draw a section (profile) of the image of the point source and evaluate its half-width FWHM on the resulting PSF curve, which is expressed in millimeters. However, a line source is more suitable for this purpose - see below "
Measuring the positional resolution of the camera ".

Two basic ways of measuring the homogeneity of the field of view of a gamma camera.
Testing and calibration of the internal homogeneity of the field of view of a scintillation camera without a collimator using a point source located at a sufficiently large distance. Measurement of the overall homogeneity of a scintillation camera with a collimator using a flat homogeneous source. Double-sided area homogeneous source 57 Co for testing the homogeneity of a 2-detector SPECT camera

2. Total homogenity - planar source
To control and test the overall ( extrinsic ) homogeneity of the field of view of the scintillation camera with collimator , planar homogeneous sources with radionuclide cobalt
57 Co (T 1/2 = 272 days; emits gamma radiation with energy 122keV) are often used . , close to 140keV 99m Tc) , which have the same radionuclide distribution density over their entire area (surface homogeneity should be better than 99%, total activity is 200-400 MBq)
- fig.3.1.1 in the middle. The dimensions of the source must ensure full coverage of the field of view of the gamma camera (the dimensions of the source should be at least 2 cm larger than the field of view) . With this source we can quickly and operatively check the homogeneity of the camera's field of view with the collimator. The advantage is that we do not have to take off the collimator and we get a picture of the overall (resulting) homogeneity, on which we can also recognize a possible defect of the collimator - mechanical deformation in the septa. Only when the homogeneity is unsatisfactory will we recalibrate the camera's photomultipliers using a 99m Tc
point source as mentioned above.

Flat cobalt <- versus ->
" box " homogeneous sources
Previously used "box" homogeneous sources filled with an aqueous solution of 99m Tc are now rarely used in larger workplaces. Their disadvantages were considerable laboriousness in filling them, risk of contamination, higher radiation load during preparation, difficult mixing of the solution to be truly homogeneous and the possibility of inhomogeneous absorption of radionuclide on the phantom walls, which can disrupt homogeneity - in other words " for a lot of money
(effort) little music "..! .. The disadvantage of cobalt 57 Co homogeneous sources is the relatively higher purchase price - with a larger number of scintigraphic examinations, however, they form only a small fraction of the price for radiopharmaceuticals ... If the workplace does not have a flat cobalt source, we recommenduse only method 1. with a point source 99m Tc (removing the collimator is much easier than filling the "box" source with a radionuclide solution) .

Fig.3.1.2. Some typical images of the homogeneity of the field of view of a camera with a surface source or a point emitter (after application of a correction matrix) .
a) Normal image of homogeneity. b) Peripheral photomultiplier failure.
Totally detuned photomultipliers or photopeak set outside the analyzer window. d) Cracked scintillation crystal.
It is discussed in more detail in 4.2, section " Technical failures of scintillation cameras "

Double-sided homogeneous sources
Previously manufactured
(until the end of the 1990s) flat homogeneous sources 57 What were one-sided : the exact homogeneity of the radiation field was guaranteed only from one - "front-front" wall, from the opposite "back" side the homogeneity was only approximate. Therefore, when measuring homogeneity in dual-head SPECT cameras, it was necessary to perform the acquisition for each detector separately. Nowadays, double-sided flat homogeneous sources are produced , which can be inserted between the heads of the SPECT camera and measure the homogeneity of both detectors at the same time - Fig.3.1.1 on the right.
Radionuclide purity of homogeneous sources?
First homogeneous area sources
57Co, which began to be produced in the late 1970s, sometimes contained radionuclide impurities 56 Co (t 1/2 71 days, E g 511, 847, 1038, 1238, 1771, 2598, 3253 keV) and 58 Co (t 1/2 79 days , E g 511, 811 keV) , which could cause artificial image inhomogeneities due to the emission of higher gamma energies. There was even a recommendation not to use "fresh" sources, but only by the age of about 6-8 months from their acquisition (due to shorter half-lives than 272 days 57 Co, the content of radionuclide impurities will be significantly reduced) .
  Now all this is pointless, as newer sources have a high radionuclide purity directly from production
(max. content of radionuclide impurities <0.1-0.2%) .
Note .: If we in the spectrometric measurements of cobalt-57 in addition to the basic power 122 + 136keV found higher 692keV gamma peak, not a radionuclide impurities but the actual legitimate gamma peak of radiouklidu - see spectrum measured 57 Co .
Inhomogeneity correction
By measuring the image of a homogeneous source, a matrix of correction coefficients g
ij can be created in a computer , which multiplies the uncorrected values ??of scintigraphic images a ij in individual pixels (i, j), thus creating a corrected image a *ij :

Fig.3.1.3. Computer correction of gamma camera image inhomogeneity.
Left: The uncorrected image and ij homogeneous source, having considerable inhomogeneity.
Middle: Matrix of correction coefficients g ij .
Right: Multiplying by correction factors produces a corrected image * a ij that is already homogeneous.
Note: Instead of the usual luminance modulation , an isometric display is used here , where the height of the elements (pixels) above the base is proportional to the number of pulses contained. The curves at the top are cross-sections , taken through the center of the images.

Virtually all disturbances and anomalies in the imaging properties of the camera are most sensitive to the homogeneity of the field of view. Therefore, regular homogeneity testing is required to ensure quality scintigraphic imaging . Testing time intervals are recommended at least once a week, with impaired stability of the camera's electronic circuits every day (individual camera types have their own recommended testing intervals) . And of course after every electronic intervention in the circuits of photomultipliers, amplifiers and ADCs. In the case of impaired homogeneity, it is necessary to adjust or recalibrate the photomultipliers (tuning) and create a new correction matrix, in the case of gross abnormalities, electronic intervention or repair.
  How the homogeneity of the camera is quantified, calibrated and corrected is given in 4.5 " Physical parameters of scintigraphy ", section "
Homogeneity (uniformity) of the camera's field of view ". Dependence of the field homogeneity on the physical conditions
The homogeneity of the gamma camera display depends (in addition to the mechanical, detection, optical and electronic properties of the device) also on a number of physical conditions and setting parameters . Hence the principles that must be followed when performing correct phantom measurements to correct homogeneity. We will illustrate them on scintigraphic images of experimental measurements:
Note: In brightness modulated images, very sharp brightness modulation is used to emphasize subtle differences - - LT 80%, UT 95%. The differences in homogeneity are in fact much milder ...
- Statistics - number of pulses in the image
If a sufficient number of accumulated pulses in the image (information density) is not achieved, statistical fluctuations can also manifest as inhomogeneity . Of course, these disturbing statistical fluctuations also affect the homogeneity correction matrix if it is accumulated with an insufficient number of pulses. It can be seen in the figure that an insufficient number of correction matrix pulses (eg 200 imp./pixel) can very adversely affect the static scattering of the corrected image and thus impair the recognizability of the lesions. The correction matrix should therefore be recorded with as many pulses as possible, min. 1000, optimally 5000 imp./pixel :

Fig.3.1.4 Influence of statistical fluctuations of the correction homogeneity matrix on the correction result. The image of a homogeneous
99m Tc source with an HR collimator (uncorrected on the left) is gradually corrected by matrices stacked with different numbers of pulses / pixels - information densities ID.

- Used collimator
Classic collimators with parallel holes HR, HS, HE
(properties of different types of collimator are described in 4.2, section " Scintigraphic collimators ") , if they do not have a mechanical defect, they do not in principle affect the homogeneity of gamma camera display. For divergent / convergent collimators the homogeneity is already somewhat different, for Pinhole they differ significantly; however, the correction of this difference is usually not performed. Just make a correction for the internal inhomogeneity of the detector .

Fig.3.1.5. Influence of collimator on homogeneity. In the upper half there are uncorrected images of the homogeneous source
99m Tc obtained with different collimators, in the lower half these images are corrected by the matrix for the collimator HR.

- Frequency of pulses during camera image acquisition
The dependence of the camera image homogeneity on the frequency of registered pulses is relatively small. A more significant effect begins to manifest itself at a registered frequency higher than about 30,000 imp./sec. And with a very high flux of gamma photons
(high activity of the measured object, order GBq) incident on the camera crystal, the camera detector and its electronic circuits are flooded , there is a shift, an increase in dead time . When recording the correction matrix, the activity of the source used should be such that the pulse frequency does not exceed about 10,000 cps.

Fig.3.1.6. Influence of pulse frequency on homogeneity.
In the upper half, there are uncorrected images of a homogeneous
99m Tc source obtained at different frequencies of registered pulses, given the activity of the source.
In the lower half, these images are corrected by a low frequency matrix (1000 cps).

- Gamma radiation energy of the used radionuclide
Scintillation in the crystal, registration of the photomultiplier and positional coding of the generated pulses depend on the energy of gamma photons. It can be seen from the images in our experimental measurements that the homogeneity of the field of view depends on the gamma energy, so it is generally recommended to correct each scintigraphic image with a matrix obtained for the same gamma radiation energy. However, small differences in energies do not have a significant effect on homogeneity, so for example images with technetium
99m Tc or gallium 67 Ga can be corrected by a matrix for cobalt 57 Co - which is commonly done (as mentioned above) .

Fig.3.1.7. Influence of gamma radiation energy on the homogeneity of the camera's field of view. In the upper part there are uncorrected images of homogeneous sources of different radionuclides with different energies of emitted gamma radiation. Below, these images are corrected by a matrix for 140 keV technetium
99m Tc.
The spectrometric "detuning" of one of the peripheral photomultipliers can be seen at the bottom left of the field of view.

- Spectrometric adjustment of the analyzer window to the photopeak - its width and symmetry
The adjustment of the analyzer window affects not only the detection efficiency and the proportion of Compton scattered radiation, but also the homogeneity of the image. With the window set asymmetrically to the total photopeak, a slightly different part of the local photopeak can be detected or cut off for individual photomultipliers , which leads to local differences in detection efficiency - image inhomogeneity . However, small differences in the spectrometric setting (up to 5%) do not have a significant effect on the homogeneity and the correction result.

Fig.3.1.8. Influence the homogeneity of the camera's field of view by setting the analyzer window. In the upper part there are uncorrected images of a homogeneous
99m Tc source with different window settings (width 20%) on the photo peak. Below, these images are corrected by a matrix for symmetrical window adjustment to 140 keV technetium 99m Tc.

- Time factor - instrument instabilities
The homogeneity of the gamma camera can also change over time due to instabilities in the detector and electronic circuits. The correction matrices must therefore be checked from time to time and renewed if necessary (disagreement).

Fig.3.1.9. Example of measuring time stability and changes in the homogeneity of the camera's field of view. The image of the homogeneous source
99m Tc is gradually corrected by a "differently old" matrix - at different times between the creation of the correction matrix and the acquisition of the image for correction.
The instability of one of the peripheral photomultipliers can be seen at the bottom left of the field of view.

Technical Note: Physical measurements of homogeneous sources and correction matrices from which the above images are derived were performed in 1975-78 (as part of the research report " Computer Data Processing in Nuclear Medicine ") on a PhoGamma HP Nuclear Chicagogamma camera from 1973. (picture .... left). The pictures show a certain detuning and instability of some photomultipliers (especially the peripheral photomultiplier at the bottom left) for a specific "piece" of the camera.
The details of the design and electronic circuits have since undergone considerable technical development. However, the basic physical and technical principles still apply.

3.2. Measurement of the spatial resolution of the gamma camera
The physical-technical parameter spatial resolution of the camera is defined and discussed in more detail in 4.5 " Physical parameters of scintigraphy - image quality and phantom measurements " of the passage " Spatial resolution ", here we present only the basic picture :
Fig.4.5.1 Spatial scintigraphic resolution - analysis of point source images that appear as "blurred" scattering circles.
More detailed discussion in 4.5, passage " Spatial resolution ".
To test the positional resolution of a gamma camera and the linearity of the image
(see below "Analysis of linearity imaging") is used to either point and a line source, or so Bar-phantoms .
For practical determination of resolution is preferable to use a line source. Compared spot has a line source is preferably that the image may lead independently multiple profiles, respectively. These profiles add - lead a wider section-profile (about 5-10 pixels), which gives a summed LSF curve with smaller statistical fluctuations (Fig.3.2.1).
The line source
is used for quantitative physical measurement of the positional resolution of the camera . be larger than approx. 0.5-1 mm *), which we fill with a radionuclide solution
(most often 99m Tc) and display using a camera with a collimator. We then draw sections - profiles - through the image of the line source and on these LSF ( Line Spread Function ) curves we determine the width at half height - FWHM , which expressed in millimeters indicates the total positional resolution of the camera for a given distance from the collimator front - Fig.3.2.1.
*) If the test point or line source fails to make a sufficiently small (thin), a quadratic correction is required for objective evaluation of the spatial resolution :
        FWHM = [FWHM m 2 - D 2 ] 1/2 ,
where FWHM m
 is the measured (uncorrected) half-width, D is the diameter or thickness of the point or line source, and FWHM is the resulting actual (corrected) resolution.
  For detailed physical analysis, the so-called modulation transfer function of the MTF camera can be calculated from the profile curve of the line source , which gives the contrast of the model cosine distribution of radioactivity depending on the spatial frequency - see " Modulation transfer functions " in the chapter " Mathematical algorithms of scintigraphy " .

Measurement of spatial resolution by a line source

We keep a suitable profile (section) by painting the line source
The half-width D 1/2 of the profile curve of the line
source indicates the positional resolution FWHM
The modulation transfer function can be calculated from the profile curve of the line source

For precise measurement of the positional resolution of gamma cameras (in planar mode), it is suitable to incorporate the line source into the following simple phantom : we attach our own thin capillary in the middle to a plexiglass plate, in which two small holes with a diameter of about 1 mm are excavated at the sides at a distance of 10 cm . Using a syringe with a thin needle, inject a column of radionuclide solution into the capillary, eg 99m Tc - this is the line source itself . We drip the same solution into the excavated wells in the same way - two point sources are createdat a precisely known distance, to determine the scale of the display. We place the phantom in the field of view of the camera equipped with the required collimator and take scintigraphic images at different distances. When evaluating, we keep transverse profiles (sections) of the image in the places of point sources - from the distances of the pixels of their maxima we determine the scale of the display , by which we multiply the horizontal coordinates in the graphs of profiles. From the profile via the image of the line source - LSF - we then determine the spatial resolution of the FWHM (or the MTF modulation transfer function ) :

Fig.3.2.2. One way to accurately measure the spatial resolution of a gamma camera.
Left: A simple phantom with a line source (capillary) and two point sources for measuring the positional resolution of a gamma camera.
Middle, right: The measurement was performed with a charge of 99m Tc at distances of 0, 5, 10, 15 and 20 centimeters from the front of the collimator of the HR camera Nucline TH. The degradation of FWHM resolution with distance can be seen in the images and profile curves.

Transmission so-called bar-phantoms are used for a simple visual assessment of the resolution (and possibly also linearity) of the camera. The bar-phantom consists of a plastic plate with a system of sealed lead absorbent strips of various widths. There are gaps between the lead strips (the width of the gaps and strips is the same) ; under the camera, the bar-phantom is placed over a semi-homogeneous source. The lead strips then absorb radiation from a homogeneous source, while gamma radiation passes freely through the gaps to the camera collimator. In the scintigraphic image, we observe which width of the strips is still distinguishable - the actual positional resolution is then approximately given by 1.75 times the narrowest still resolved stripe):

Gamma camera resolution testing using bar-phantom.
Transmission bar-phantom with stripes
in 4 segments
An image of a bar-phantom with stripes in
4 segments
An image of a bar-phantom with horizontal stripes of different widths.

3.3. Gamma Camera Display Linearity Analysis
Linear line sources
are used to analyze the linearity of the image or image distortion
(as well as the resolution - see below) . Deviations from the linear shape can be assessed on the image of such a line source (4.5 " Physical parameters of scintigraphy ", passage " Linearity of the camera's field of view ") . We display line sources in different places of the camera's field of view and also in different angles, especially in two mutually perpendicular directions. In principle, the above-mentioned bar phantoms can also be used to evaluate linearity . However, the most perfect phantom for analyzing the linearity of scintillation camera imaging is Cartesian  linear grid . It consists of a regular network of parallel and perpendicular thin tubes (hoses with a diameter of about 1 mm), placed at equidistant distances (eg 2 cm). The tubes are filled with a solution of the appropriate radionuclide, usually 99m Tc, with a suitable activity (total approx. 100MBq).

Fig.3.3.1 Cartesian linear lattice filled with
99m Tc radionuclide solution .

The phantom is placed in the field of view of the camera, the basic measurement is close to the front of the collimator. For different types of collimator (especially for convergent or Pinhole), the display is also performed at different distances to evaluate the effect of distance on the display scale and on the linearity or distortion of the image:

Fig.3.3.2 Scintigraphic images of a linear rectangular grid with different types of collimators, taken at different distances.

In a collimator with parallel holes (such as LE HR, left) we get a linear representation of the grid everywhere , while for greater distances from the front of the collimator, the spatial resolution deteriorates (blurred grid). With a convergent collimator (such as a SmartZoom with a convergent center portion), the image of the center portion increases with increasing distance . With the Fan Beam collimator (which is convergent in the transverse direction, parallel in the axial direction), the grids increase only in the transverse direction with increasing eye distance , they remain the same in the axial direction . The Pinhole collimator shows the most significant dependence on the object distance : we get the image many times close to the foreheadmagnified , with increasing distance the zoom decreases and for distances above about 20cm the image is already reduced . The pictures also show a general trend of deteriorating resolution (and thus contrast in the image) with the distance from the front of the collimator.
   If the lattice phantom is precisely executed and the inner diameter of the tubes does not exceed 1 mm, it can be advantageously used to measure local spatial resolution in different places of the field: at the desired location of the image height to peak (FWHM), which is the resolution at a given location. Possibly. the regional modulation transfer function MTF for a given image location can also be calculated .
We assembled the first such linear grid at our nuclear medicine office in Ostrava-Poruba in 1975. It consisted of a long thin plastic tube with an inner diameter of 1 mm, which was intertwined with equidistant grooves (in two perpendicular directions), cut in a polystyrene board. Using a syringe, it was filled with a solution of the desired radionuclide, especially
99m Tc. It had dimensions of 30 30cm and was used for Pho Gamma 3, HP cameras with parallel collimators, Pinhole convergent and divergent, Slant Hole, and then for MB9100,9102 cameras . The current phantom (in Figure 3.3.1) is made with the same technology, but has larger dimensions - 52 times46cm to cover the field of view of current SPECT cameras .

3.4. Tomographic phantoms for SPECT, PET, CT
For physical analysis of basic imaging properties of tomographic cameras, point or linear sources can be used, similar to planar scintigraphy. Special cylindrical phantoms , mostly of the Jasczak or Venstra type, are used for visual assessment and testing of tomographic imaging properties of SPECT and PET cameras in clinical scintigraphy . These are Plexiglas cylinders, which are filled with a solution of the appropriate radionuclide
(for SPECT mostly 99m Tc, for PET 18 F) . Inside they contain a systemrods and beads of various sizes, where the radioactivity does not get and which therefore simulate " cold lesions " - Fig.3.4.1 above. Another variant is refillable containers , into which we can inject a solution of a radioindicator with a suitable specific activity to simulate " hot lesions " and their quantification ( SUV , see below) .

Phantom Jasczak folded (top)
and unfolded (bottom)
Tomographic SPECT images of the upper and lower parts of the Jasczak phantom
  68 Ge / 68 Ga cylindrical phantom as a source of annihilation gamma radiation 511keV for testing and calibration of PET cameras
                     Fig.3.4.1 Two types of tomographic phantoms for SPECT and PET

We perform a tomographic acquisition and on the images of cross sections, after reconstruction, then we monitor the resolution of the lesions depending on the size. We can test the influence of various aspects of acquisition (number of accumulated impulses, number of projections, distances, displacements of the center of rotation, etc.) and evaluation (reconstruction algorithm - rear projection or iterative reconstruction, used filters) on lesion resolution or artefacts. Phantoms with refillable lesions can also be used to calibrate the SUV parameter (described in 4.2, section " Scintigraphic image quality and detectability of lesions "), especially for PET imaging. In addition, at the end of the phantom, there is still free space filled with a homogeneous distribution of the radionuclide, which is used to test tomographic homogeneity . As already mentioned above, special anatomical inserts imitating the structures of the brain, heart, liver, kidneys and the like can also be placed in the space of the cylindrical phantom .
   In addition to phantoms operatively filled at the workplace with a solution of radionuclides
(mostly short-term - 99m Tc, 18 F) , solid (closed) phantoms containing suitably distributed long- lived radionuclides are also used for gamma camera testing . Most often they are planar homogeneous sources with a radionuclide57 Co (their use was described above in the section " Testing and calibration of scintillation camera image homogeneity "). Closed so-calledgermanium phantoms, emitting annihilation gamma radiation of 511 kV,are often used for calibration and testing of PET cameras. They are filled with the parent radionuclide 68 Ge , which with a half-life of 271 days is converted by electron capture to the short-termpositronradionuclide gallium 68 Ga . The source of 511 kV annihilation radiation, detected by a PET camera, is daughter gallium-68(formed by the interaction of positrons emitted by68Ga with the electrons of the material), maternal germanium does not participate in radiation. Point and line sources (mostly also germanium) are rarely used . The positron radionuclide 22 Na with a half-life of 2.6 years can also be used for PET phantoms (however, a certain disadvantage here is the high proportion of hard gamma radiation 1274keV) .
Simple improvised phantom for measuring the imaging properties of SPECT and PET cameras
For measuring the basic imaging properties of PET cameras - resolution, detection sensitivity, astigmatism - I made a very simple " phantom " - Fig.3.4.2. From a 5 cm thick polystyrene foam board for door insulation, I cut a rectangle 20
x 70 cm at the cottage at the weekend , to which I glued small plastic conical cuvettes *) to the center r = 0 and at distances of 10, 20, 30 and 34 cm . At the KNM workplace, I then injected small droplets of fluoride radionuclide solution 18 F onto the conical bottom of the cuvettes using a micropipette about exactly the same activities (for a specific measurement in the picture it was 1.85MBq) - this created radially distributed point sources . I then placed this "phantom" exactly symmetrically inside the detection ring of the PET camera.
*) This "primitive" solution was forced by the circumstances: During the general reconstruction of our workplace in 2011, we lost a relatively well-equipped electronic and mechanical workshop with all devices, mechanical tools (lathe, vertical drill, scissors, ...) and tools. . Therefore, the mechanically perfect design was already difficult for me to access. However, even this simple solution was completely satisfactory from a functional point of view ...     
A simple improvised phantom for measuring the imaging properties of PET and SPECT cameras.
   After the acquisition (10 min.), a PET image of these point sources was created , on which we evaluated the spatial resolution and detection efficiency depending on the position r in the PET detection ring using ROI and profiles - cf. 4.3, part " Positron emission tomography of PET ", passage " Spatial resolution of PET ". The effect of radial " astigmatism " of PET imaging for greater distances from the center of the PET ring was also well seen there (it is discussed in the mentioned passage "Spatial resolution of PET" in 4.3) and also a slight decrease in the detection efficiency towards the peripheral areas.

Fig.3.4.3 PET images of point sources 18 F located at different distances r from the center of the detection ring. By analyzing the ROI and profile curves with these images, the values ??of detection efficiency h and spatial resolution FWHM were measured (we measured on a GE Discovery PET camera at KNM FN Ostrava) .
At the last peripheral point source at a distance of r = 34 cm, part of its image was already cut off by the edge of the field of view.

For comparison, we placed a similar phantom (0-20cm, according to the typical dimensions of SPECT, eg chest) , with point sources filled with 99m Tc , between the circulating SPECT cameras . In the reconstructed cross-sectional tomographic image, we guided the profiles through point source images in an analogous manner and determined the spatial resolution and detection efficiency for individual distances r from the center of rotation :
SPECT images of 99m Tc point sources located at different distances r from the gantry center.
   The worse spatial resolution of SPECT imaging is due to the relatively large distance from the front of the collimator (22 cm) - it is typical for SPECT chest; in SPECT brain (distance of lesions from collimators approx. 10 cm) the resolution is approx. 9-10 mm. Overall, a somewhat better resolution for PET is given by electronic collimation (4.3, passage " Spatial resolution of PET ") . This is also the reason for the many times better detection efficiency (sensitivity) of PET .

Testing and correction of the center of rotation SPECT cameras
When tomographic SPECT scintigraphy orbiting detectors, cameras weighing hundreds of kilos around the object to be examined. If the bearings and arms in which the detectors are mounted in the gantry show mechanical play , gravity forces cause the detectors to swing and shift, so that the rotation does not occur exactly around a fixed axis - the center of rotation ( COR ) moves . These unwanted mechanical shifts and fluctuations degrade the quality of the reconstructed tomographic images
(and could possibly lead to artifacts). For basic testing of the center of rotation, we place a point source in the field of view of the camera and start its SPECT scintigraphy. The relevant program then evaluates the deviations of the resulting pattern from the circular motion and determines the displacements of the axis of rotation depending on the instantaneous angle of the detector. Another place with the risk of unwanted mechanical movements of the camera detectors during rotation are their own bearings on which the detectors are mounted. To test these events. axial fluctuationsa single point source is not enough - several (eg 3-5) point sources distributed in specified places of the field of view are used. Appropriate sensing and correction are done using special acquisition software for modern SPECT cameras. For cameras with two or more detectors ("heads"), a more complex measuring procedure is used, called MHR (Multiple Head Registration) , where concurrency is also measured and calibrated - matching the position of images from individual detectors (cf. the section " Geometric matching of CT images " below) with SPECT and PET images ") ; a larger number of point sources is used here, approx. 10.
Based on the above measurements, the appropriate correction coefficients are stored for each angle of rotation.
   , which, when acquiring patient studies, perform appropriate image shifts in the X and Y directions so that deviations from the center of rotation and axial oscillations of the detectors are eliminated .
Note: In PET positron emission tomography with stationary detection rings without mechanical movements, of course, this unfavorable effect does not exist and corrections of the center of rotation and MHR are not performed .

Tomographic phantoms for CT
In addition to testing and calibrating the scintigraphic part (as described above), the SPECT / CT and PET / CT hybrid systems also measure and test the imaging properties of the CT part - homogeneity, image contrast, positional resolution, noise. Basic measurements for daily check of the correct function of the X-ray machine and opposite detectors (" check-up ") are performed without a phantom - " through the air ". Further measurements and testing of specific CT imaging properties are then performed using cylindrical water and plate phantoms
(there are several types recommended for different types of CT devices) . These phantoms contain a part homogeneously filled with water, in the next part there are suitable absorbent structures - plastic rollers and strips of various sizes, or metal wires (approx. 0.1 mm tungsten) - fig.3.4.5. Phantoms are inserted between the X-ray machine and the detectors into the center of the CT, transmitted by transmission, the transverse sections reconstructed, and the densities of the individual parts are displayed and quantified using Hounsfield units .
   The homogeneity and noise test is performed using a basic part of a cylindrical phantom filled only with water. Positional resolution and contract CT images are measured on images of absorption rollers and strips (different distances) located in other parts of the phantom, respectively. using LSF and MTF from wire images. The new CT devices have special programs for evaluating the image quality.

Example of two types of tomographic phantoms for testing CT imaging properties.

Geometric alignment of CT images with SPECT and PET images
When merging functional scintigraphic SPECT or PET images with anatomical CT images, it is important that the structures shown in both modalities are geometrically overlapped - displayed in the same image location (image
fusion, hybrid tomographic systems ). For the process of this geometric alignment of images, the not very apt name of registration or normalization of images is sometimes used ... Scinticraphic + is used  for this calibration of precise alignment of the positional concurrency of the displayed SPECT <--> CT or PET <--> CT structures on hybrid instruments . CT imaging of point sources filled with mixtures of radionuclide ( 99m Tc or 18 F) and contrast agent . The radioactive substance in these samples is "seen" by the SPECT or PET gamma camera, the contrast agent is "seen" by CT. In the simplest case, it improvises 2ml. syringes, optimally placed in special cuvettes filled with about 0.1-0.2 ml. a radiolabelter solution mixed with a contrast agent. Several (usually 5-10) such point sources are used, regularly distributed in the field of view of the device - they are inserted into the holes of a simple plate phantom.
   This system of point sources, containing both radionuclide and contrast agent, is simultaneously - in-line - displays scintigraphically on a gamma camera (SPECT or PET) and density on CT. The calibration program uses cross-correlation of scintigraphic and CT images to determine the map of the respective mutual displacements - scale and affine transformations, so that all point sources on SPECT or PET and CT images are exactly geometrically covered . This map is then automatically used for clinical SPECT / CT or PET / CT image fusions.

4. Dynamic phantoms
These are phantoms that model various dynamic processes in the body, such as heartbeat, blood flow in blood vessels, filtering activity of the kidneys or liver, lung breathing, swallowing of the esophagus. Mathematical evaluation of dynamic scintigraphic studies provides a number of quantitative parameters of the dynamics of investigated biological processes, whose actual values a prioriunknown and which may be affected and distorted by some influences, such as statistical fluctuations, geometric influences, absorption of radiation in the tissue, inaccurate definition of areas of interest, inadequacy of the mathematical model used, individual variability in patients, etc. For the exact analysis of these influences and for the development of methods and algorithms for the calculation of quantitative parameters, the need arose for the most complex types of phantoms, which are dynamic phantoms. These phantoms model the temporal dynamics of changes in the distribution of the radioindicator in the relevant organs and their parts. The advantage is that the actual values ??of the parameters of this dynamics are known - they are provided by the construction of the phantom, or can be precisely set .

Such a dynamic phantom can be useful in three stages of research and application work in the field of nuclear medicine:

  1. During the development of a new measuring method, selection and tuning of the correct mathematical model and algorithm for evaluating the measured data.
  2. When calibrating the measuring method and determining the correct calibration or regression coefficients when evaluating the measured data.
  3. When authentication and "training" has been developed and used methods.

Here we will briefly describe several dynamic phantoms that we developed (and constructed either on our own or with the help of a specialist workshop) at the Department of Nuclear Medicine in Ostrava during research and development work in the field of physical-mathematical analysis and computer evaluation of scintigraphic studies.
   Our simplest dynamic phantom was the phantom of the periodic event realized in 1973 by a rotating turntable carrying a radioactive source (see below). In 1976, when developing algorithms for mathematical analysis of radionuclide ventriculography (including the geometric method of calculating the absolute volume of the heart chamber), we used various pulsating balloons connected by a tube to a large-volume calibrated syringe, the piston of which we moved manually - see picture:

We then used the experience gained with this simple phantom to develop our motor-driven flexible dynamic phantom of cardiac activity, which allows us to model both ventricular pulsation and central hemodynamics. Finally, we describe our phantom for modeling the dynamics of the swallowing act of the esophagus, including reflux and antiperistaltics.

Rotating phantom of the periodic process
A characteristic feature of cardiac activity is its periodicity . The model of the periodic event is a circular motion (oscillating motion is not very suitable here) . Our first dynamic phantom consisted of a gramophone , on the plate of which point sources were placed - containers with a 99m Tc solution . A turntable with rotating point sources was placed in the field of view of the scintillation camera. This phantom played an important role in the development of the methodology of scintigraphic scanning and analysis of fast periodic events - it was at a time when scintigraphic systems could not yet scan gated studies. An aperture was attached to the edge of the turntable interrupting the light beam into the phototransistor - this simulated the R-wave of the ECG . We designed electronic circuits (in cooperation with Ing. Dubrok), which implemented these "R-wave" pulses from the "ECG" into the scintigraphic data flow and developed software that reconstructed the data from LIST-mode into scintigraphic images and constructed a phase dynamic study of one cycle synchronously composed of many running cycles of the periodic process. Thus, the methodology of gated radionuclide ventriculography was developed on the CLINCOM device and put into practice .

Dynamic scintigraphy of the phantom of a fast periodic event - 3 point sources of 99m Tc placed on a turntable rotating at a speed of 78 rpm.
a) Record of the scintigraphic study in LIST mode; E - gating pulses from the phototransistor simulating the R-wave RKG.
b, c, d, e) Images reconstructed into Frame-mode with different time resolution.
f) Definition of the area of ??interest in the summary picture.
g) Common resource flow curves - Significant statistical fluctuations are seen.
h) The same curves created by the synchronous composition of 60 common cycles - fluctuations significantly reduced.

Flexible dynamic phantom of cardiac activity for radionuclide ventriculography and angiocardiography
Dynamic scintigraphy of cardiac activity analyzes two important aspects : 

During research and development work on mathematical analysis and computer evaluation of dynamic scintigraphic studies of radionuclide ventriculography and radiocardiography, we have developed and in cooperation with the mechanical workshop of VŽKG Ostrava made a comprehensive dynamic phantom of cardiac activity - Fig.1. This phantom can be used both for radionuclide ventriculography, where it models heart chamber pulsation, and for bolus angiocardiography, where it models cardiac pumping activity. All parameters, such as ejection fraction, heart rate volume, residual volume, end-diastolic volume, ejection and filling rates, heart rate, minute heart volume, phase shifts of "gating" pulse transmission, can be continuously changed and precisely set in a wide range. This phantom thus enables a very complex analysis and verification of radionuclide cardiological methods and their precise calibration .
Phantom drive unit
The basis of the entire phantom is a calibrated working cylinder (glass), in which the piston moves . The displacement of the piston is mechanically secured via a rod (connecting rod) from a disk with an eccentrically located joint driven via a gearbox by an electric motor - Fig. 1a. The eccentricity of the joint position on the drive disk can be precisely adjusted, which defines the stroke of the piston - pulse volume . The frequency can also be changed continuously using the motor speed controller movement of the piston in the cylinder - heart rate . Phase sync pulses ("ECG gain pulses") are sensed photoelectrically. A diaphragm is mounted at a suitable location around the circumference of the rotating drive disk , which obscures the flow of light from the bulb as it passes around the phototransistor. The generated electrical impulses are conducted via a normal cardiomonitor to the computer - they simulate the R-wave of the ECG . The diaphragm can be freely moved around the circumference of the drive disk and thus change and precisely set the phase of the periodic process in which the gating pulses will be transmitted - it can be used, among other things, for Fourier phase-amplitude analysis . The entire working cylinder is located on the support, by the displacement of which any residual volume can be set continuously, regardless of the size of the piston stroke.

The described phantom of cardiac activity can work in two basic modes :

  1. Pulsing balloon mode with closed system (Fig.2).
  2. Flow piston pump mode (Fig.4).

The drive unit and the control part of the phantom, including the working cylinder, are common to both modes.

Fig.1a. Detail of the drive part of the phantom with a drive disk and a scale for adjusting the stroke volume and with an aperture and a photoresistor emitting synchronization pulses.

Fig.1b. A general view of the phantom in a mode modeling the pumping activity of the heart.

Phantom of the pulsating heart chamber
To model the pulsation of the heart chamber during gated ventriculography
(4.9.4 " Radionuclide gated ventriculography ") , the phantom works in the arrangement according to Fig.2. In this mode, the phantom is suitable for testing the calculation of ejection fraction , cardiac volume, ejection and filling rates, time intervals of significant cycle phases, Fourier phase-amplitude pulsation analysis, verification and calibration of methods for calculating absolute ventricular volume, some aspects of background and correction on tissue background.

Fig.2. Dynamic phantom of heart activity in pulsating balloon mode. In this variant, the phantom is used to model the pulsating heart chamber during radionuclide ventriculography.

A rubber balloon is placed in the field of view of the camera, which is connected by a hose to the working cylinder of the phantom. The closed system (working cylinder - balloon) is filled with an appropriate amount of radioactive solution of 99m Tc of equilibrium concentration, the volume and activity of which can be easily changed via the filling valve. The motor-driven oscillating movement of the piston to the left and right leads to the periodic filling and emptying of the radioactive jaw from the balloon - the balloon thus pulsates similarly to the heart chamber. By changing the stroke volume of the piston, its speed of movement and the initial position, it is possible to easily and accurately set any value of ejection fraction, heart rate and residual volume, heart rate, and thus minute heart volume (here imaginary). The geometric configuration also results in the value of the maximum and average ejection and filling speed.
The balloon pulsates in the aquarium below the level of the radioactive solution, which represents the absorbing tissue and the tissue background. The depth of the balloon placement and the specific activity of the background solution can be easily changed.
   The thus pulsating balloon is scanned by a scintillation camera, and with the help of gating pulses emitted by the diaphragm and the phototransistor, a phase dynamic study of one cycle is synchronously composed of a large number of running cycles as in ventriculography. In the mathematical analysis and computer evaluation of this study, they are then compared calculated values ??of dynamic parameters with their actual values ??set on the phantom. Part of such an evaluation of a phantom study of ventriculography using the VENTR program in our OSTGAM system on the GAMMA-11 instrument is shown in Fig. 3; good agreement was recorded between the calculated and actual parameters.

Fig.3. Part of the results of a complex evaluation of phantom radionuclide ventriculography of a pulsating balloon using the VENTR program on a Gamma-11 device.
Left: Scintigraphic images of the balloon in end-diastole and end-systole and table of actual set values of phantom parameters.
Right: The balloon volume curve, generated from the activity time curve after background correction and volume normalization using the geometrically calculated absolute ED of the balloon volume, together with the calculated dynamic parameters.

Phantom of Central Hemodynamics
Figure 4 schematically shows an arrangement in which the phantom models the pumping activity of the heart in central hemodynamics. This mode is mainly used for analysis and testing of bolus angiocardiography - calculation of cardiac volume, transits, volumes
(4.9.4 " Dynamic radiocardiography ") .

Fig.4. Arrangement of a dynamic phantom for modeling the pumping activity of the heart for radionuclide angiocardiography. Note: The scintillation camera actually "looks" at the working cylinder and the dilution vessel perpendicular to the drawing.

The "left ventricle" in this case is a working cylinder, which is located together with another dilution vessel (representing the "right ventricle + lungs") in the field of view of the camera. From the storage tank (reservoir with a volume of about 5 liters - can be changed) representing the total blood volume, when moving the piston to the right draws (sucks) through the dilution vessel and one-way suction valve water into the working cylinder, when the piston moves opposite, the suction valve closes and water it is pushed through the tube back into the storage container via the open second one-way valve. The valves alternately close and open, so the phantom works here as a piston pump with continuously adjustable parameters. The product of the stroke and heart rate fraction indicates the flow rate of the pump, ie "minute heart volume".
   A radioactive bolus (approximately 100MBq 99mTc in 1 ml) is injected into the rubber suction tube at the appropriate time and a dynamic radiocardiography (RKG) study is started. The bolus quickly proceeds first to the dilution vessel ("right chamber"), from where it is, partially diluted, sucked into the working cylinder and from there - already completely diluted - expelled in parts into a storage vessel. The time course of radioactivity in the dilution vessel and in the working cylinder is thus the same as in the right and left heart chambers during bolus radiocardiography: a sharp increase with bolus arrival, a slight exponential decrease due to dilution leakage, then systemic recirculation of the radioindicator concentration of radioactivity in the system.

Fig.5. After injecting a bolus into the phantom suction tube, we see in the sequence images the gradual passage of radioactivity through the dilution vessel and working cylinder, from where it is gradually expelled into the storage vessel with constant dilution, after which part of the radio indicator returns by system recirculation. significantly weaker). In the next course, the radio indicator is homogeneously diluted throughout the system (these images are no longer captured here).

The resulting phantom dynamic study is then processed by the RKG program for the evaluation of radiocardiography. The gamma-functions and exponential functions are interleaved with the curves of the time course of radioactivity in the respective areas of interest (bolus arrival, dilution vessel, working cylinder), which represent the passage and dilution of the radioactive bolus in individual parts of the pumping system, transit times, volumes and cardiac output. "similar to angoicardiography; the calculated parameter values are confronted with the actual parameters set on the phantom. In Fig.6. part of the evaluation of such a bolus phantom study using our RKG program is shown. The calculated parameters again agreed very well with the actual parameters of the phantom.

Fig.6. Verification of the operation of the RKG program for complex evaluation of bolus radiocardiography using phantom measurement in the arrangement according to Fig.4.
Left: Images of some stages of the passage of radioactivity through the phantom; below is a table of actual phantom set parameters.
Right: Part of the results of the evaluation of the phantom study by the RKG program (version on Gamma-11).

With the described phantom, a number of measurements were performed both in the pulsating balloon arrangement of Fig. 2 and in the flow pump mode of Fig. 4. The phantom has played an important role in the development of methods and algorithms for dynamic methods of nuclear cardiology, as well as in the creation of appropriate software that is still successfully used (VENTR and RKG programs).
In the pulsating balloon mode, the agreement of all parameters calculated in radionuclide ventriculography (
VENTR program ) was verified and an accurate calibration of the geometric method for determining the end-diastolic volume of the heart chamber was performed; a proportional geometric-analytical method for determining the absolute volume of the heart chamber was also developed and verified.
In the flow pump mode, the dynamics of flow and dilution were analyzed and the accuracy of the determination of minute heart volume and various aspects of combined radiocardiography + ventriculography were verified.
Another planned addition of a "short-circuit" tube with adjustable flow, connected between the left and right ventricles, unfortunately did not take place - it would be possible to test methods for quantification of intracardiac short circuits.
   In comparison with the phantoms developed so far, the described phantom has an original concept and the advantage of greater flexibility and more complex use. Even some later phantoms
(such as the "Vangerbild Cardiac Phantom" Amersham) did not reach such complexity, even though they were mechanically made professionally.
Colleagues from the Department of Nuclear Medicine in Olomouc also made a very nicely mechanically executed phantom of a pulsating balloon and also performed a number of measurements with it.

4.3 Phantom measurement dynamics esophageal
For the purpose of development of the methods of mathematical analysis and evaluation of complex dynamic scintigraphy of esophagus and stomach (4.9.3 " scintigraphy esophagus ") , as well as for testing the correctness of the program results OESOGAST, we designed a simple dynamic phantom modeling passage of tracer marked bite by esophagus with the possibility of simulating anomalies, including antiperistaltics or gastro-oesophageal reflux.
   The phantom is shown in the figure on the left. It consists of a longer shaft-helix (length about 60 cm), into the thread of which fits a housing with a holder for gamma radiation sources. The shaft is built vertically in the field of view of the scintillation camera and rotates using an electric motor with the possibility of changing the direction of rotation. When the shaft rotates, due to the threads, the housing with the radioactive emitter also moves in the field of view of the camera in the vertical direction up or down - depending on the direction of rotation of the electric motor. The downward movement simulates the passage of the bite through the esophagus during the swallowing act, the upward movement of any reflux or antiperistaltics.

Dynamic phantom of a radiolabel passage through the esophagus. Simulation of reflux using two emitters.

During our own phantom measurements, we place a source of radiation g - a syringe or an ampoule with a radionuclide 99mTc with an activity of about 50 MBq -
in the holder (connected to the case on the helix) . We can use one or two ampoules (see below). Move the case to the upper position, then start the electric motor in the appropriate direction and start a dynamic scintigraphic study. A number of phantom measurements have been performed with various modifications of this arrangement. The basic phantom measurement models the normal case
, ie fast and uniform   passage of the radioindicator through the esophagus during the swallowing act. Place one container with a radio indicator in the holder set above the upper edge of the camera's field of view, start the downward movement and start a scintigraphic study of this uniform (linear) movement of the radio indicator. We then stop the movement of the radio indicator at a suitable stage (it represents the attainment of the stomach), but let the scintigraphic study run for a while. The result of the evaluation of the normal phantom study program OESOGAST is shown in Fig .A . The curves of the passage of the radio indicator of the upper middle and lower part of the "esophagus" have the same height and width, while the calculated transit times agree with the actual times of movement of the source along the helix, measured by the stopwatch. Also, the transport functions and fused images M and her exactly linear in shape.

A. Normal steady motion B. Movement with retention simulation C. Motion with antiperistaltic simulation

The result of another phantom measurement is shown in FIG. It differs from the previous normal experiment in that at a certain stage of the downward movement of the radio indicator, the movement was temporarily stopped for 2 sec. This anomalous movement of the radioindicator is again very well visible in the results of the computer evaluation - the curve of the middle part of the "esophagus" is extended by 2 sec. transit time, condensed image and tra n sports functions also faithfully reflect the trajectory of movement.
   During the next phantom measurement, the direction of rotation of the motor was temporarily changed during the downward movement, so that the radio indicator changed after 2 sec. he moved upwards, after which, after switching again, the movement continued evenly downwards. This uniform "swallowing" movement with temporary return is also faithfully represented and quantified on a computer processing of a scintigraphic study - Fig.C.
Another modification of the phantom measurement coincided with the normal experiment of even swallowing for most of its course, but at the end during the stationary phase of the radioindicator resting in the "stomach" the motor was started again in the opposite direction
(upward movement), so that the radio indicator returned to the "esophagus", after changing direction, which it dropped back to the "stomach" position. The result of the evaluation can be seen in Fig. D as a simulation of complete gastro-oesophageal reflux.

D. Movement with simulation g.-e. reflux 100% E. Simulation of gastro.-esophagus. reflux 50%

The last type of phantom experiment uses two g radiation sources located in a phantom holder (see figure above right). The course of the measurement is similar to that of a normal experiment - uniform movement of two closely spaced sources (they are not distinguished in scintigraphic images, there is one source with total activity). However, after reaching the lower resting position of the "stomach", we move one of the sources upwards for a while (manually to the defined position), leave it here for a while and then lower it again into the case next to the other source. This simulates gastro-oesophageal refluxparts of the radio indicator from the "stomach" to the "esophagus". The result of the evaluation can be seen in Fig. E, where it can be clearly seen from the condensed image how part of the radio indicator "swung" temporarily from the "stomach" to the lower part of the "esophagus"; Quantitative analysis gave a value of 51.5% at reflux in good agreement with reality (both sources j e have the same activity, so that the actual "reflux" was 50%).
Phantom measurements of this kind are original and have not yet been reported in the literature (known and available to us). It was very useful in the development of individual algorithms for mathematical analysis of radioindicator transport through the esophagus (analysis of curves, construction of transport function and condensed image) and verification of the correctness of the evaluation results by the OESOGAST program
(Dynamic scintigraphy of the esophagus and stomach).
  Inspired by our work, they constructed a similar phantom of esophageal dynamics at the Department of Nuclear Medicine of the Army Hospital in Krakow. Instead of a threaded shaft, they used a toothed belt, the phantom looks very good - congratulations!

Scintigraphy   Computer evaluation

Nuclear physics and physics of ionizing radiation
Nuclear and radiation physics Radiation detection and spectrometry Radiation applications
With cintigraphy Computer evaluation of scintigraphy Radiation protection
Gravity, black holes and space   -   time physics Anthropic principle or cosmic God
AstroNuclPhysics Nuclear Physics - Astrophysics - Cosmology - Philosophy

Vojtech Ullmann