X-ray physics notes curriculum
Fundamentals of radiation
The X-ray machine
Production of X-rays
Interaction of radiation with matter
X-ray detection and image formation
Image quality (current module)
Radiation safety in X-ray imaging
Fluoroscopy
Mammography
This section links together several key ideas from earlier modules (attenuation, contrast, and noise) to explain how scatter degrades image quality and how CNR quantifies the effect.
Scattered radiation is one of the main causes of image degradation in X-ray imaging.
While the primary beam carries useful anatomical information, scatter adds unwanted background signal, reducing contrast and obscuring low-contrast structures.
The Contrast-to-Noise Ratio (CNR) provides a practical way to quantify how well two tissues can be distinguished in the presence of noise and scatter.
What Is Scattered Radiation?
When X-rays interact with matter, not all photons are absorbed, many are deflected (scattered) in new directions, primarily due to the Compton effect. We’ve covered this in detail earlier.
Key features
- Scatter intensity increases with patient thickness, field size, and kVp.
- Most scatter is forward-directed, reaching the detector and adding unwanted exposure.
- Scatter carries no useful spatial information about tissue structure.
Result:
→ Increased background signal.
→ Decreased image contrast.
→ Slight increase in noise.
Quantifying Scatter: The Scatter-to-Primary Ratio (SPR)
- Primary radiation → carries anatomical contrast (difference in transmitted intensity).
- Scatter → adds a roughly uniform “fog” to all areas.
SPR = Iscatter / Iprimary
Iscatter = Scatter intensity, Iprimary = Primary beam intensity
- Typical values:
- Extremities: 0.1–0.3
- Abdomen: 0.6–1.0
- Chest: 0.4–0.8
A higher SPR = more scatter reaching the detector = lower contrast and lower CNR.
Contrast-to-Noise Ratio (CNR)
While contrast resolution describes how small a signal difference exists, the CNR describes how visible that difference remains in the presence of noise:
CNR = ∣S1−S2∣ / σnoise
Where:
- S1,S2 = mean signal levels from two tissues or regions.
- σnoise = standard deviation of noise.
High CNR → subtle contrast differences are easily seen.
Low CNR → noise obscures contrast differences.
Relationship to SNR
- SNR quantifies image noise globally.
- CNR quantifies detectability between two adjacent regions.
- Theoretically, a system can have good SNR but poor CNR if scatter dominates and erases local contrast.
Factors Affecting CNR
| Parameter | Effect on CNR | Explanation |
|---|---|---|
| Scatter | ↓ | Adds uniform background signal → reduces numerator |
| Noise | ↓ | Increases denominator (σnoise) |
| kVp | Mixed | Higher kVp → more scatter (↓ contrast), higher photon flux (↑ SNR), depends on mA |
| mAs (dose) | ↑ | More photons → higher SNR → higher CNR |
| Grids / collimation | ↑ | Remove scatter → restore contrast |
| Patient thickness | ↓ | More scatter and attenuation → lower CNR |
| Pixel size | ↑ | Larger pixels improve SNR (↑ CNR) but may reduce resolution |
Scatter-Reduction Techniques (summary)
| Technique | Mechanism | Effect on CNR / Dose |
|---|---|---|
| Beam collimation | Limits field size → fewer scatter interactions | ↑ CNR, ↓ dose |
| Anti-scatter grids | Absorb obliquely scattered photons | ↑ CNR, ↑ dose (requires more mAs) |
| Air gap technique | Increases OID → scatter diverges and misses detector | ↑ CNR, slight ↑ dose |
| Compression (mammography) | Reduces tissue thickness | ↓ SPR, ↑ CNR, ↓ dose |
| Lower kVp (within reason) | Reduces Compton scatter fraction | ↑ CNR, but ↑ dose |
| Use of shielding / filtration | Removes low-energy photons that add dose but not contrast | ↑ dose efficiency |
A quick note on anti-scatter grids
These are important as the following comes up often in exams. The lead strips are aligned with the primary beam. Scattered radiation reaches the detector at an oblique angle and will be attenuated by the lead strips before reaching the detector. However, some of the primary X-rays will also be attenuated and an increase in mAs may be needed to maintain detector exposure.
- Structure: Alternating lead strips and radiolucent spacers.
- Grid ratio (h/D): height of lead strips / distance between them.
- Higher ratio → better scatter cleanup → higher CNR.
- But requires higher mAs → increases dose.
- Typical ratios: 8:1 for general radiography, 12–16:1 for high-detail work.
- Bucky factor: ratio of exposures with vs without grid → quantifies dose increase.
Key Takeaways and Exam Tips
- Scatter adds uniform background exposure → reduces contrast and CNR.
- CNR = (signal difference) / (noise) → determines visibility of low-contrast detail.
- High CNR requires high SNR and low scatter.
- Scatter reduction techniques (grids, collimation, air gap, compression) improve CNR but may increase dose.
- Common exam question: “Explain how scatter affects image quality and describe methods used to reduce it.”
Up Next
Next, we’ll move on to Artefacts in Digital Imaging, covering the common causes, appearances, and correction methods for artefacts in CR and DR systems.