07 September 2016

Lecture #3: INSTRUMENTATION QUALITY ASSURANCE



  
Quality assurance refers to a system designed to identify problems, propose solutions and monitor results to achieve the desired level of performance

Quality control refers to the testing designed to identify equipment problems.

THE SCINTILLATION CAMERA OR GAMMA CAMERA

Characteristics of an Ideal Gamma Camera

1.      Field Uniformity

Field uniformity is the ability of a scintillation camera to produce a uniform image when the source provides a uniform distribution of photons over the detector. This has to be performed on a daily basis.

Two types of field uniformity testing

a.      Intrinsic testing monitors the condition of the sodium iodide crystal and electronics associated with the scintillation camera detector and is performed without a collimator attached to the detector.

b.      Extrinsic testing assesses the instrument as it is used clinically so collimator is placed in it.

Comparison of Intrinsic and Extrinsic Scintillation Camera Quality Control

                                                Intrinsic                                 Extrinsic

Collimator                             None                                      Low energy
                                                                                                Parallel–hole

Radiation source                  Tc99m point source             Tc99m liquid flood
                                                Tc99m liquid flood              Co57 solid sheet
                                                Co57 solid sheet

Source distance                    point source placed                         flood source placed
                                                4 – 5 crystal diameters         directly on face of
                                                away from detector              collimator
                                   
                                                flood source placed on
                                                crystal face

Causes of field non–uniformities

a.      Misadjusted photopeak
b.      Damaged crystal
c.       Collimator defect
d.     “Mistuned” photomultiplier tube
e.      Photomultiplier tube malfunction
f.        Radioactive contamination on crystal or collimator
g.      Scratched, dirty CRT screen or multiformatter lenses
h.      Film processing artifacts

2.      Spatial resolutions

Spatial resolution is the ability of a scintillation camera to reproduce small differences in radionuclide concentration in closely spaced areas. Clinically, resolution affects the ability to visualize small defects. This can be performed intrinsically or extrinsically using a bar or multihole phantom.

a.      Hine–Duley Phantom

The Hine–Duley phantom consists of lead bars and spaces of three different widths; the bars are embedded in plastic. Since the resolution is not measured equally over all areas with this phantom, multiple images must be acquired to assess resolution for the entire detector area.

b.      Four quadrant Bar Phantom

The four–quadrant bar phantom contains spaces and lead bars of four different widths. The spaces and bars in each quadrant are of equal width. Four images, acquired at 90o to one another are necessary to monitor resolution over the entire crystal.


c.       Parallel–line Equal Space (PLES) Phantom

All lead bars and spaces are of equal width in the PLES phantom. Two images acquired at 90o to one another are necessary to monitor resolution over the entire crystal.


d.     Orthogonal–hole phantom

The orthogonal–hole phantom consists of a lead sheet containing holes of equal diameter arranged at right angles to one another. Only one image is needed to determine resolution over all areas of the detector.

            Factors affecting scintillation camera spatial resolution

a.      Photopeak adjustment
b.      PHA window width
c.       Source–detector distance
d.     Gamma ray energy
e.      Collimator type

3.      Spatial Linearity

Spatial linearity is the ability of a scintillation camera to produce a uniform image with straight lines corresponding to straight lines in a phantom. Clinically, it is the accurate portrayal of a true organ shape. Linearity can be assessed along with resolution by examining the straightness of a set of parallel bars or, in the case of the orthogonal–hole phantom, the straightness of parallel rows of holes.

4.      Sensitivity

Sensitivity is the ability to detect ionizing events in a sodium iodide crystal expressed in counts per second per microcurie (cps/ µCi). It can be measured simultaneously when uniformity is performed if the activity of the source and the room background is known. The activity in the source must be carefully measured and corrected for decay, then the count rate obtained per unit activity can be calculated.

            Sensitivity     =          (counts – background) ÷ time
                                                            activity of source

                                    =          cps / µCi or cpm / µCi

Sensitivity can be affected by incorrect energy settings, incorrect collimation, if extrinsic sensitivity is performed, or improper detector–source distance.


DOSE CALIBRATOR

Quality control tests for dose calibrator

1.      Constancy

Constancy testing determines the reproducibility of measurements of a source of known activity from day to day. A long lived standard is measured at each of the commonly used radionuclide settings. Each measurement is compared to the calculated (decay–corrected) value for the standard at that setting on that day.

a.      Count a long–lived standard, such as Co57 or Cs 137 of known activity on the appropriate dose calibrator setting

b.      Obtain  a background at the same setting to determine net activity

c.       Correct the known activity for decay and determine the ± 10% range of this value

d.     The measured value should be within 10% of the decay–corrected activity

e.      Using the same reference source, repeat the measurement for all commonly used radionuclide settings.

f.        Again correcting for background and decay, determine the ± 10% range for each calculated value. Compare each measured value to the range of calculated values for that radionuclide setting.

g.      Each measured value should be within ± 10% of the decay–corrected value at each setting.

CO 57 has a half life 271 days
Cs 137 has a half life 30 years

According to the NRC, if the constancy error is greater than 10%, the instrument should be repaired or replaced.

2.      Activity Linearity

A linearity test ascertains how accurately a dose calibrator measures activities over a wide range, from millicurie to microcurie amounts. The test involves assaying a source with a relatively short half–life over a period of several days and comparing the predicted and measured results.

a.      Assay an aliquot of Tc99m 2–3 hour interval over 3 more days. According to the NRC, the starting activity should approximate the highest activity that will be administered to a patient. The starting activity must be assayed until it reaches 30 µCi

b.      Correct all measurements for background

c.       Calculate the predicted activity for each time the sample was measured.

d.     Plot both the calculated and measured activities on semi–log paper (activity versus time)

e.      Measured activities should be within ± 10% of calculated activities

According to NRC, if linearity errors exceed 10%, mathematically corrected dose calibrator readings should be calculated for activities greater than 10 µCi.

An alternate method of performing activity linearity test is to use the commercially available system “Calicheck” or “Lineator.” These systems consist of a series of lead tubes that absorb Tc99m photons by a known amount. After the source is assayed initially, it is measured several more times after each of the lead tubes is placed in sequence around the source. Measurements of the shielded source simulate radioactive decay. This method saves time and decreases personnel exposure. The source used to perform the test may also be used in patients after linearity data collection.

3.      Accuracy

Accuracy testing assesses the ability of a dose calibrator to accurately measure the activity of radionuclides of different gamma energies. According to NRC, at least two different standards, one of which must have photon energy between 100 and 500 keV, must be assayed.
Accuracy testing is performed as follows:

a.      Assay each standard in the dose calibrator three times at the appropriate radionuclide setting. Measure the background at each setting.

b.      For each standard, average the three readings and subtract the background to obtain the net activity.

c.       Correct the known reference activity for decay and determine the ± 10% range of the decay–corrected value.

According to NRC, if the accuracy error is greater than 10%, the dose calibrator should be repaired or replaced.

4.      Geometric variations

Geometric variation determines the effect of sample volume and configuration on the measurement of sample’s activity.

For example, 20 mCi contained in a 1–ml volume syringe and 20 mCi contained in a 10 ml volume glass vial may yield significantly different measurements when assayed in the same dose calibration

a.      Assay 1–2 mCi Tc99m in a minimal volume (1 – 2 drops) in a 20 – 30 ml vial. Obtain the background reading.

b.      Increase the volume in the vial stepwise by adding water. The volumes should be those commonly assayed.

c.       Assay the vial after each addition of water and obtain a background reading.

d.     Correct each measurement for background and calculate a correction factor for each volume.

% correction factor               =          true activity              
                                                            measured activity

e.      Plot the correction factors against the volumes on linear graph paper

f.        After assaying a vial of activity in the dose calibrator, calculate the true activity using appropriate correction factor

True activity       =          measured activity    x   correction factor

g.      A similar procedure can be used for various size syringes

According to the NRC, correction factors should be calculated for any geometric variations causing measurements to vary by more than 10% from the true activity.

SCINTILLATION SPECTROMETER

A scintillation spectrometer is an instrument, such as an uptake probe or scintillation well counting system that uses pulse–height analysis to determine which pulses will be counted.

1.      Calibration of a scintillation spectrometer involves determining the appropriate operating high voltage and amplifier gain.

a.      Operating voltage is the high voltage applied to the dynodes of the photomultiplier tube (PMT). The high voltage can be adjusted so that arbitrary units on the pulse–height analyzer dials, upper and lower level discriminators (ULD and LLD), corresponds to meaningful energy (voltage) units. The operating voltage is determined in the following manner

(1)   Use a monoenergetic gamma source of known energy like Cs137

(2)   Set a narrow window around the photopeak energy by adjusting the ULD and LLD of the pulse–height analyzer. For Cs137, the LLD is 652 and ULD is 672 in 3% window.

(3)   Beginning with the lowest voltage setting, count the source for 1 minute at each voltage setting, increasing the voltage until the maximum count rate is observed.

(4)   Plot counts per minute versus energy (voltage) on linear graph paper. The voltage setting at which the highest count rate is obtained is the operating voltage. The units on the pulse–height analyzer dials are now calibrated for 0–1000 keV. The window can be moved up or down to accept pulses from the photopeaks of other radionuclides.

(5)   Note gain settings. Calibration results are valid only on that gain settings.

Increasing the voltage moves the entire gamma spectrum to the right. When the photopeak moves into the preset window, the maximum counting rate is observed.

b.      Amplifier Gain

Adjusting the amplifier gain will permit counting of very high or very low radionuclide energies. When the amplifier gain is used to alter pulse height and shift the gamma spectrum, one unit on the ULD and LLD dials is no longer equivalent to 1keV.

Types of Gain Control

(1)   True Gain Control – an increase in gain produces an increase in pulse– height with a corresponding decrease in the energy range.

(2)   Inverse Gain Control – an increase in gain produces a corresponding decrease in pulse–height but a corresponding increase in the energy range.

2.      Energy Resolution – is the conversion from light to electrical energy and is indicated by the photopeak. This can be checked using Cs137.

a.      Using the proper operating voltage and a narrow window, measure the counts for a preset time at 10 keV increments around the photopeak.

b.      Plot counts per 10 keV interval versus the midpoint of the counting interval.

c.       Calculate the full width at half maximum (FWHM) as follows:

(1)   Divide the maximum photopeak count by 2

(2)   Draw a line across the photopeak at 50% count level determined by Step a.

(3)   Drop perpendicular lines to the x – axis where the line in step b intersects the two edges of the photopeak. The distance between the 2 perpendicular lines (in energy units) is the FWHM

(4)   Calculate the % energy resolution as follows:

% Energy resolution            =                      FWHM (keV)                x100                                                           Midpoint of photopeak (keV)

A typical range for Cs137 is 8% – 12%. The energy resolution will degrade, that is, the value will become larger as the crystal ages or if a problem develops with the amplifier or photomultiplier tube.

3.      Chi Square Test

The Chi square test is a method to check for random errors greater than those that would be predicted. Random errors affect the reproducibility (or precision) or measurements. These are always present in radiation counting because of the random nature of radioactive decay.

a.      Using a long lived standard (Cs137 or Co57), collect a series of measurements (20 – 50), each present time.

b.      Compute the mean average value.

c.       Calculate the chi square value using the formula:

X2        =          Σ(Ni – N)2
                             N
d.     Using the x2 table, find the probability of p. The p value denotes the probability that random variations observed in a series would equal or exceed the calculated x2 value.

Acceptable p value is between 0.1 and 0.9, values greater than 0.9 or less than 0.1 indicate that the variations in the measurements do not match the expected and the instrument should be checked.

SURVEY METER

Survey meter Quality Control

1.      Reference Check are performed by measuring a long lived source, using constant geometry

2.      Calibration







No comments: