Gamma Camera: Structure & Function Revision Notes | A Level AQA Physics

1. Why the Gamma Camera Matters

A gamma camera does something remarkable: it lets clinicians see inside a living patient without surgery, by detecting gamma rays emitted from a radioactive tracer that has been injected into the bloodstream. The image it produces reveals not just anatomy but function — how well a kidney filters, whether a tumour is metabolically active, how blood flows through the heart.

To understand how this works at the physics level, you need to trace the journey of a single gamma photon from the moment it leaves a nucleus inside the patient's body to the moment it becomes a dot on a medical image. That journey passes through three critical components: the collimator, the scintillator crystal, and the photomultiplier tube array. Each one has a precise job, and each one is examinable.

2. Overview: The Structure of a Gamma Camera

Picture the gamma camera as a large, flat detector head held close to the patient's body. Working from the patient outward, the layers are: the collimator (closest to the patient), then the scintillator crystal, then an array of photomultiplier tubes pressed against the back of the crystal, and finally the position logic circuitry that converts the PMT signals into a two-dimensional image.

• Collimator: lead plate with parallel channels — selects direction of incoming gamma photons
• Scintillator crystal: NaI(Tl) — converts gamma photons into visible light flashes
• PMT array: converts each light flash into an amplified electrical pulse
• Position logic / Anger circuit: calculates where in the crystal the flash occurred
• Computer display: builds up the image pixel by pixel from thousands of detected events

Each component is necessary. Remove any one of them and the camera either stops detecting radiation, loses all spatial information, or produces a signal too weak to measure. The design is a chain — and understanding why each link exists is exactly what AQA examiners test.

3. The Collimator: Selecting Direction

Gamma rays are emitted in all directions from every nucleus that decays inside the patient. Without some way of filtering them, the detector would receive photons from every angle simultaneously and the resulting image would be a blurred, useless smear. The collimator solves this problem.

Think of the collimator like a honeycomb of drinking straws made of lead. If you look straight down a straw, you see through it. If you tilt the straw, the wall blocks your view. Each channel in the collimator acts the same way: it only passes gamma photons that are travelling in the correct direction. This means each point on the crystal corresponds to a specific direction — and therefore a specific location — in the patient's body.

There is a genuine trade-off here. Making the channels narrower and longer improves spatial resolution (sharper image) but reduces the fraction of photons that get through, lowering sensitivity. Wider, shorter channels let more photons through but blur the image. This is a classic design compromise that AQA may ask you to evaluate.

4. The Scintillator Crystal: Converting Gamma to Light

Once a gamma photon passes through the collimator, it enters the scintillator crystal. The crystal used in almost all clinical gamma cameras is sodium iodide doped with thallium, written as NaI(Tl). The thallium dopant creates energy levels within the crystal's band structure that allow visible light to be emitted when the crystal absorbs ionising radiation.

When a gamma photon enters the NaI(Tl) crystal, it interacts with the crystal lattice — primarily through the photoelectric effect and Compton scattering — and deposits its energy. This energy excites electrons within the crystal. As those electrons return to lower energy states, they emit photons of visible light. A single gamma photon from a typical medical radionuclide (such as technetium-99m, which emits 140 keV gamma rays) produces a burst of approximately 4,000–6,000 visible light photons. This burst is called a scintillation.

The crystal must be large enough to intercept a useful fraction of the gamma photons that pass through the collimator, and thick enough to stop most of them (rather than letting them pass straight through). NaI(Tl) crystals in clinical cameras are typically around 9.5–12.7 mm thick and can be up to 50 cm in diameter. The crystal is hermetically sealed because sodium iodide is hygroscopic — it absorbs moisture from the air, which would degrade its optical properties.

5. The Photomultiplier Tube: Amplifying the Signal

The scintillation flash produced by a single gamma photon is extremely faint — far too weak to measure directly with conventional electronics. The photomultiplier tube (PMT) solves this by converting the light flash into an electrical signal and then amplifying that signal by a factor of millions, all within a single vacuum tube.

A gamma camera uses an array of PMTs — typically 37 to 91 of them — arranged in a hexagonal pattern across the back face of the crystal. Each PMT is a cylindrical vacuum tube, roughly 5–8 cm in diameter, with its flat face pressed against the crystal (via an optical coupling material to minimise light loss at the interface).

5a. Inside a Photomultiplier Tube: Stage by Stage

The internal structure of a PMT has three main regions: the photocathode, the dynode chain, and the anode. Each region has a specific function in the amplification chain.

• Photocathode: a thin layer of photosensitive material (e.g. caesium-antimony compound) coated on the inside of the glass entrance window. Absorbs visible light photons and emits photoelectrons via the photoelectric effect.
• Focusing electrode: an electrostatic lens that directs the emitted photoelectrons towards the first dynode.
• Dynode chain: a series of electrodes (typically 10–12 dynodes) held at successively higher positive potentials. Each dynode emits several secondary electrons for every electron that strikes it — this is the multiplication stage.
• Anode: the final electrode, held at the highest positive potential. Collects all the electrons produced by the dynode chain and outputs a measurable electrical pulse.

Worked example: Tracing a Signal Through a Photomultiplier Tube

Describe, in sequence, how a single visible light photon entering a PMT produces a large electrical pulse at the anode. Assume the dynode chain has 10 stages and each dynode emits 4 secondary electrons per incident electron.

1. 1 visible light photon → 1 photoelectron (via photoelectric effect) — The photocathode material has a low work function, so even low-energy visible photons can eject electrons. The quantum efficiency (fraction of photons that successfully eject an electron) is typically 20–30% for a good PMT photocathode.
2. The focusing electrode and the potential difference between photocathode and dynode 1 accelerate the photoelectron. — Each successive dynode is held at a higher positive potential (typically 100–150 V higher than the previous one). The electron gains kinetic energy as it accelerates towards the positively charged dynode.
3. Dynode 1: 1 electron in → 4 electrons out\nDynode 2: 4 electrons in → 16 electrons out\nDynode 3: 16 electrons in → 64 electrons out — When an accelerated electron strikes a dynode, it has enough kinetic energy to eject multiple secondary electrons from the dynode surface. With a multiplication factor of 4 per dynode, the number of electrons grows as a geometric sequence: $4^n$ after $n$ dynodes.
4. $4^{10} = 4 \times 4 \times 4 \times 4 \times 4 \times 4 \times 4 \times 4 \times 4 \times 4$\n$= 16 \times 16 \times 16 \times 16 \times 16$\n$= 256 \times 256 \times 256$\n$= 65536 \times 256$\n$= 1{,}048{,}576 \approx 10^6$ electrons — Starting from a single photoelectron, 10 stages of multiplication by a factor of 4 produce approximately one million electrons. This is the gain of the PMT per photoelectron. Real PMTs with 10 dynodes and a multiplication factor of around 4–6 per dynode achieve gains of $10^6$ to $10^7$. Note that because the photocathode quantum efficiency is only ~20–30%, not every incident visible photon produces a photoelectron, so the overall efficiency from incident photon to anode output is lower than the gain figure alone suggests.
5. $\approx 10^6$ electrons arrive at the anode → electrical pulse output — The anode collects the electron cascade and the resulting charge pulse is passed to the preamplifier and signal processing electronics. The pulse height (amplitude) is proportional to the energy deposited in the crystal by the original gamma photon.

Answer: One photoelectron from the photocathode is amplified to approximately $10^6$ electrons at the anode after 10 dynode stages (each with a multiplication factor of 4), giving a PMT gain of approximately $4^{10} \approx 10^6$ per photoelectron. The overall efficiency from an incident visible photon to anode output also depends on the photocathode quantum efficiency (~20–30%).

The key point to carry forward is that the PMT gain is multiplicative at every dynode stage. This is why the amplification is so dramatic — it is not additive but exponential. A gain of $10^6$ means a signal that was originally unmeasurably small becomes a clear, processable electrical pulse.

6. Position Determination: The Anger Logic Circuit

With the PMT array producing electrical pulses, the next challenge is working out where in the crystal the original scintillation occurred. This is essential for building a spatially accurate image. The solution, developed by Hal Anger in the late 1950s (published 1958), is elegant: use the relative pulse heights from neighbouring PMTs to calculate the position by weighted averaging.

When a scintillation occurs at a particular point in the crystal, the light spreads out in all directions. PMTs directly above the flash receive more light and produce larger pulses; PMTs further away receive less light and produce smaller pulses. The position logic circuit (Anger circuit) computes the X and Y coordinates of the flash using a weighted centroid calculation across all PMT outputs. The result is a single (X, Y) coordinate for each detected gamma photon.

Simultaneously, the circuit sums all the PMT outputs to get the total pulse height, which is proportional to the energy of the original gamma photon. This total energy signal is used for pulse height analysis (also called energy windowing): only events whose total pulse height falls within a set energy window around the expected gamma energy are accepted. This rejects Compton-scattered photons, which have lost energy and would otherwise degrade image contrast.

7. Putting It All Together: The Full Detection Chain

Now that each component has been examined individually, it is worth tracing the complete detection chain from patient to image. This is the kind of sequential description that AQA 'describe' questions require — and getting the order right, with the correct physics at each stage, is what earns the marks.

Worked example: The Complete Gamma Camera Detection Sequence

Describe the sequence of events by which a gamma photon emitted inside a patient's body produces a contribution to the final radionuclide image.

1. A radionuclide (e.g. technetium-99m) in the patient's tissue undergoes nuclear decay and emits a gamma photon in a random direction. — Technetium-99m is the most widely used radionuclide in clinical imaging. It emits 140 keV gamma photons, which are energetic enough to escape the body but low enough to be stopped efficiently by the NaI(Tl) crystal.
2. The gamma photon either passes through a parallel channel in the lead collimator (if travelling perpendicular to the detector face) or is absorbed by the lead septa. — Only photons travelling in the correct direction reach the crystal. This is what gives the image its spatial information — each point on the crystal maps to a specific location in the patient.
3. The gamma photon is absorbed by the crystal (primarily via the photoelectric effect). The deposited energy excites electrons, which emit a burst of several thousand visible light photons as they return to lower energy states. — The number of visible photons produced is proportional to the energy of the gamma photon. For a 140 keV gamma photon in NaI(Tl), approximately 4,000–5,000 visible photons are produced per event.
4. Visible photons spread through the crystal and enter the PMTs. Each PMT's photocathode converts incoming photons to photoelectrons, which are then amplified through the dynode chain to produce electrical pulses. — PMTs closer to the scintillation site receive more light and produce larger pulses. The variation in pulse height across the array is the raw data used for position calculation.
5. The position logic circuit computes the weighted centroid of all PMT outputs → (X, Y) coordinates. It also sums all outputs → total pulse height (proportional to gamma energy). — The energy signal is passed through a pulse height analyser. Events outside the accepted energy window (e.g. Compton-scattered photons with reduced energy) are rejected to maintain image contrast.
6. Each accepted event adds one count to the pixel at position (X, Y) on the display. After thousands or millions of events, a two-dimensional map of tracer distribution builds up. — Regions of high tracer uptake produce many counts and appear bright on the image. Regions of low uptake appear dark. The image reflects the physiological function of the organ being studied.

Answer: Gamma photon → collimator (direction selection) → NaI(Tl) crystal (gamma to visible light) → PMT array (light to amplified electrical pulses) → Anger circuit (position and energy calculation) → image display (count accumulation at X, Y coordinates).

8. Exam Technique and Common Mistakes

AQA questions on the gamma camera tend to use 'describe' and 'explain' command words. A 'describe' answer should state what happens at each stage in sequence, without needing to justify why. An 'explain' answer must include the underlying physics — for example, stating that the dynode emits secondary electrons because the incident electron has sufficient kinetic energy to eject electrons from the dynode surface.

For questions about the PMT, examiners want to see the correct sequence: photocathode → dynodes → anode, with the photoelectric effect named at the photocathode stage and secondary emission named at the dynode stage. Omitting either of these named processes typically costs marks. Similarly, for the collimator, you must state that it absorbs photons travelling at an angle — not merely that it 'filters' them.

Every component in the gamma camera chain — from the lead collimator that selects photon direction, through the NaI(Tl) crystal that converts gamma energy to visible light, to the PMT array that amplifies that light into a measurable electrical signal, and finally the Anger circuit that localises each event — is essential to producing a clinically useful radionuclide image. Understanding the physics of each stage, and being able to explain the role of each component with precise terminology, is the foundation of AQA exam success on this topic.