Gamma cameras

Question 1

What are the major components of a gamma camera?

Answer 1

• imaging collimator (use to define the direction of the detected gamma rays; controls which gamma rays are accepted to form a projected image of the distribution on the surface of the crystal)
• large area NaI(TI) scintillation crystal
• light guide
• array of photomultiplier tubes
• pre-amplifiers/ADCs
• pulse arithmetic circuit aka position logic circuit (takes signals from PMTs and determine the X-Y location of each scintillation event by using the weighted average of PMT signals)

Question 2

Give 3 reasons why gamma rays are the preferred emissions for radionuclide imaging

Answer 2

1. They have high enough energies (80-500keV, 511keV) to be able to penetrate through soft tissue, so can be detected from within deep-lying organs
2. They can be stopped efficiently by dense scintillators
3. They can be adequately shielded with reasonable thicknesses of lead

Question 3

Define static imaging

Answer 3

Imaging an unchanging radionuclide distribution

Question 4

Define dynamic imaging

Answer 4

Imaging changes in the radionuclide distribution over time; allows for physiologic information acquisition, e.g. rate of tracer uptake, rate of clearance from an organ of interest

Gated images e.g. of heart also possible via synchronisation to electrocardiogram signals

Question 5

What are the physical properties of an NaI(Tl) detector crystal?

Answer 5

• a single crystal
• large area (up to 60x40 cm)
• rectangular
• thickness ~6-12.5mm

Question 6

How is the NaI(Tl) detector crystal housed?

Answer 6

• Surrounded by a highly reflective material (e.g. Ti02) to maximise light output
• Hermetically sealed (air-tight) inside a thin Aluminium casing to protect it from moisture
• An optical glass window on the back surface of the casing permits scintillation light to reach the PMTs

Question 7

What is the purpose of the scintillator (crystal)?

Answer 7

To absorb gamma radiation and convert it to light photons.

The signal (intensity of the light produced) is proportional to the energy deposited (by the gamma rays) within the crystal.

Question 8

What is a scintillation detector?

Answer 8

A scintillation detector is a scintillator AND a device (e.g. a PMT) that converts the light into an electrical signal.

Scintillators are materials that emit visible light/UV after the interaction of ionising radiation with the material (electrons raised to excited energy level, ultimately fall back to lower energy states with the emission of VIS/UV light)

Question 9

Explain the scintillation process

in a NaI(Tl) scintillation crystal

Answer 9

• Gamma ray interacts in the crystal
• Gives up its energy in full (photoelectric absorption) or partially (Compton scattering)
• Secondary electron is produced, travels through the crystal causing ionisation of nearby atoms
• Each ionised atom loses an electron
• Electron is excited from valence to conduction band of the crystal energy levels, leaving a hole
• Falling back to valance band by non-radiative transmissions (producing heat)
• OR: electron-hole pair migrates to activation centre, falls back via radiative transfer
• Scintillation light emitted
• The number of light photons emitted is proportional to the energy deposited by the gamma ray in the crystal (crystal is a good energy discriminator)

The Tl doping provides activation centres to ensure immediate light emission (fluorescence)

1 gamma photon becomes 20-30 photons per keV

The number of scintillations produced is counted by the PMTs

Question 10

What are the advantages of NaI(Tl) scintillator crystals?

Answer 10

1. High gamma ray stopping efficiency thanks to its high density (3.67g/cm^3)
2. High probability of photoelectric absorption rather than Compton scattering; atomic number of iodine Z = 53, effective Z = 50, makes it an efficient absorber of gamma rays < 300keV (a.k.a. diagnostic energies) where the predominant mode of interaction is photoelectric absorption
3. High conversion efficiency of absorbed energy to light (aka an efficient scintillator; has high output of photons per keV at room temperature = good energy resolution (i.e. 1 visible light photon per ~30 eV of radiation energy absorbed))
4. Short scintillation time/excited state lifetime (230 ns) = allows high count-rates (the system only detects one gamma photon at a time; to many photons hitting the camera is bad, it tops out and can’t distinguish between the 1st and 2nd gamma rays in terms of timing)
5. It is transparent to its own scintillation emissions = low loss of scintillation light from self-absorption, therefore can construct large detectors without significant loss
6. A 9mm thick crystal will absorb 84% of 140keV photons but only 13% of 511keV annihilation photons (gamma cameras are only useful at energies < 300keV; if the crystal is too thin it won’t absorb all the gamma rays coming through the collimator)
7. It can be grown relatively inexpensively in large plates - advantageous for imaging detectors
8. Scintillation light wavelength is well-matched to peak response of the photocathode in the PMT

Question 11

What are the disadvantages of NaI(Tl) scintillators?

Answer 11

1. Fragile - need to avoid mechanical and thermal stresses (replacing the crystal is a big job, want to avoid it as long as possible; need to avoid rapid temperature changes and remember to put the collimator back on because if the building heating is switched off overnight, the crystal might crack!)
   NB: crystal fractures don’t necessarily destroy its use but they do create opacities that reduce the amount of scintillation light reaching the photocathode.
2. Hygroscopic - hermetic sealing is required (the Aluminium case) as exposure to moisture or humid atmosphere causes yellowish surface discolouration which impairs light transmission to the PMT.
3. At higher gamma ray energies > 250 keV, Compton interaction is dominant, so larger volumes of NaI(TI) are reqired for adequate detection efficiency.

Question 12

NaI(TI) crystal detector thickness is a trade-off between…

Answer 12

• detection efficiency (increased thickness = increased stopping efficiency)
• intrinsic spatial resolution (increased thickness = deteriorating spatial resolution; more potential multiple scatters; fewer photons for tubes’ nearest event)

Question 13

What is the approximate NaI(TI) crystal thickness in a gamma camera?

Answer 13

General purpose gamma cameras - thickness = 9.5mm

Can have adequate detection efficiency with lower energy gamma emitters e.g. 99m-Tc or 201-Tl using crystal thickness = 6mm

Question 14

How are PMTs housed?

Answer 14

• The array of PMTs is coupled to the back face of the crystal with a silicone-based adhesive or grease to minimise internal reflections at the interface
• PMTs can be encased in thin magnetic shields to prevent changes in the gain (caused by changes in gamma camera orientation WRT earth’s magnetic field). The focusing of the electron beam from one dynode to the next can be affected by external magnetic fields
• Hermetically and light-tight sealed in glass, evacuated (keep out moisture and extraneous light and for mechanical protection)
• Electrical connections to the dynodes, photocathode, and anode are made through pins in the tube

PMTs are ultrasensitive to magnetic fields.

Question 15

How does a PMT work?

Answer 15

PMTs are electronic tubes that produced a pulse of electrical current when stimulated by very weak signals.

1. Source of light (visible photons) impinge on glass entrance window
2. Window is coated with a photoemissive substance (e.g. CsSb, ejects electrons when struck by photons of visible light) a.k.a. the semi-transparent photocathode turns the photons into photoelectrons of energy E = hf - w [eV] with conversion efficiency = QE
3. A series of focusing grids direct the photoelectrons towards dynodes
4. Dynodes (positively charged metal plates, 200-400 V relative to the photocathode) attracts the photoelectrons.
5. Dynode coating has high secondary emission characteristics (e.g. CsSb again); high-speed photoelectrons striking its surface causes several secondary electrons to be ejected. Electron multiplication factor is typically x3 - x6 per dynode but depends on energy of initial photoelectron (which is determined by the voltage difference between the dynode and photocathode)
6. Secondary electrons are attracted to the 2nd dynode (voltage is 50-150 V higher than 1st dynode); electron multiplication process repeats through many dynode stages (~9-12 is standard)
7. Shower of electrons is collected at the anode

Question 16

What is the quantum efficiency QE of photoemission at the photocathode?

Inside a PMT

Answer 16

QE = (number of photoelectrons) / (number of incident photons)

Typically ~20-30%

~30% of light emitted by the crystal actually reaches the cathode of the PMT

Question 17

What are dynodes?

(in a PMT)

Answer 17

Metal plates maintained at a positive voltage, typically 200 - 400 V, relative to the photocathode.

Question 18

How many dynodes are present in a PMT and what are their electron multiplication factors?

Answer 18

• 9 to 12 dynodes is typical
• electron multiplication factors of x3 to x6 per dyanode is typical

Even for relatively weak light signals, a relatively large pulse of current is produced when the tube is stimulated.

total electron multiplication factor is very large, e.g. 6^10 (~6x10^7) for a 10-stage tube with an average multiplication factor of 6 at each dynode

Question 19

What is the amount of current a PMT produces proportional to?

Answer 19

Amount of current produced is proportional to the intensity of the light signal incident on the photocathode, and therefore also proportional to the amount of energy deposited in the crystal by the radiation event)

Question 20

What sort of voltage supply do the PMTs require?

Answer 20

1. high voltage supply (if the tube has 10 dynodes with the first one being at +300V WRT the photocathode, each with additional +100V increments up to the anode, the voltage supply needed is 300 + 9(100) + 100 = 1300V
2. stable voltage supply (electron multiplication factor is very sensitive to dynode voltage changes, e.g. a 1% increase in voltage supplied to the tube = 10% increase in current collected at anode!)

Stability is particularly importance when pulse sizes are measured, e.g. in pulse-heigh spectrometry to determine gamma ray energies.

Question 21

What is the purpose of the PMTs in a gamma camera?

Answer 21

To create an electronic signal from the light emitted by the scintillation crystal when a gamma ray interacts with it. Also to amplify the signal (via the dynodes)

Question 22

How is the signal acquired by the PMTs analysed?

(basics)

Answer 22

PMTs get different signals depending on their location over the crystal.

A gamma photon hits the crystal at some location. The PMT behind the interaction location will acquire the highest amount of signal. The signal strength decreases away from this particular PMT.

Sum the PMT signals to get the gamma energy - for 99m-Tc, events are calibrated to 140keV, i.e. 70keV gamma ray would produce half the maximum signal).
Relative signals of PMTs allow event localisation

Question 23

What does the gamma ray spectrum of a Tc-99m point source look like?

Answer 23

Gamma ray spectrum shows the counts at different energies.

• Scatter up to Compton edge ~45keV
• Backscatter and Pb x-ray peak ~80keV
• Iodine escape peak ~120keV
• 99m-Tc peak at 140.5keV (energy resolution is the FWHM of this peak)

Question 24

What does the Tc-99m gamma ray spectrum from a patient look like?

Answer 24

A patient is not a point source! Uptake of the radiopharmaceutical is across all different areas of the body. Thanks to scatter, events look like they’ve come from different locations in the patient, even if they haven’t.

The 99m-Tc photopeak is still at 140.5keV, but there is lots of Compton scatter in the patient, causing the LHS of the peak to have a tail.

Question 25

What does the gamma ray spectrum of Cs-137 look like?

Answer 25

Relative count rate vs photon energy graph
* 37keV barium x-rays peak (counts ~60)
* 184keV backscatter ‘peak’ (counts ~37)
* 478keV Compton edge (counts ~30)
* 662keV Cs-137 photopeak (counts ~100)

Question 26

Why is the inorganic NaI crystal doped with impurities (TI)?

Answer 26

NaI crystals will scintillate in their pure state but only at liquid nitrogen temperatures.
They can become efficient scintillators at room temperatures if small amounts of thallium are added.
Tl creates energy levels other than those inherence to the NaI crystal; provides a way for excited electrons to recombine with valence-band holes radiatively (by light emission) rather than non-radiatively (thermally)
Different band gap = different electron energy = different wavelength of light emitted during de-excitation/electron-hole recombination; ensures light cannot be absorbed by crystal (self-transparency property desirable in scintillators)

Inorganic scintillators are crystalline solids - they scintillate because of characteristics of their crystal structure i.e. individual atoms/molecules do not scintillate.

Question 27

What are impurity activated scintillator crystals?

Answer 27

Crystals with atoms of other elements are added to them.
The impurities cause disturbances in the normal crystal matrix structure; they are responsible for the scintillation effect.

Impurity atoms in the crystal matrix are sometimes called “activator centres”

Question 28

What are the key considerations when choosing an inorganic scintillator?

Answer 28

1. Sufficient stopping power: ability of the scintillator to stop high energy gamma rays (>100keV)
2. Decay time: determines the precision with which the time of gamma ray interactions in the scintillator can be determined. Faster light production within the scintillator = faster decay time = better timing precision. Decay time is also the limiting factor in how many gamma ray interactions a detector can process per unit time; two interactions can be unambiguously detected if separated by ~2-3 decay times, otherwise events “pile up” on each other which leads to dead time. Faster decay time = handle higher event rates
3. Efficiency of gamma ray to visible light photon conversion: a.k.a. photon yield, important for determining the precision with which the energy of the interacting gamma ray can be determined - particularly important when trying ti distinguish between Compton scattered and unscattered gamma rays. Higher photon yield = higher positioning accuracy of the imaging system (limited number of scintillation photons are shared among the PMTs to determine the interaction’s location)
4. Refractive index: determines how efficiently scintillation light can be coupled from the crystal into the PMT. PMT window is glass (n~1.5) so for best light transmission into PMT with minimal internal reflection, scintillator needs refractive index as close to 1.5 as possible. A lot of scintillators have n>1.5, which is one of the reasons only a fraction of the scintillation light produced actually reaches the PMT
5. Emission spectrum of light produced by scintillator matches quantum efficiency of PMT/photodetector used to convert scintillation light into an electronic pulse: scintillators with peak light emission in the 350-475nm range are optimal for use with PMTs; the transmission of scintillation light through the PMT glass entrance window should be considered too (glasses are efficient absorbers of UV with wavelenghts &laquo_space;350nm)

Decay time T, maximum event rate that scintillator can handle is ~T/2

Question 29

Why does only a fraction of the produced scintillation light reach the PMT?

Answer 29

Poor coupling of the scintillation light into the PMT due to refractive index mismatch between the scintillation material and the glass entrance window of the PMT (glass n ~1.5, scintillation n often significantly higher)

QE ~ 20-30%

Question 30

How thick can the shielding on the gamma camera detector head be?

Answer 30

For adequate attenuation of the highest gamma ray energies likely to be encountered, the thickness might need to be up to a few centimeters of lead.

Question 31

Why do collimators reduce the efficiency of a gamma camera?

Answer 31

They reject a significant amount of gamma rays emitted by the patient; they have to select rays travelling along the direction of the holes only in order to localise the events and make an image.

Question 32

Why is fluorescence prefered to phosphorescence in the scintillation process?

Answer 32

Fluorescents = immediate light emission
Phosphorescence = delayed light emission

Fluorescence occurs within < 1us of gamma ray absorption; light pulses are very brief; detector can deal with high count rates

Question 33

What are the ideal properties of a scintillation material?

Answer 33

• high efficiency for stopping gamma rays
• high probability of photoelectric absorption (instead of Compton scatter)
• high conversion efficiency (absorbed energy –> light)
• transparent to its own scintillation light
• wavelength of emitted light matches PMT response
• short scintillation time
• mechanically robust