Reference
Modalities at a glance
One table, one set of columns, and the source of every row printed beside it. Where a figure is not published by a primary source, the cell says so instead of guessing.
This is the comparison table the rest of the site refers back to. Every row was read at its source on 6 September 2026, and the source is named in the row itself. Where a primary source does not publish a figure, the cell says so rather than carrying an estimate.
The first division is the one the UK Health Security Agency draws in its public guidance Medical imaging: what you need to know, published on 22 September 2022. X-ray imaging, which includes general radiography, dental radiography, fluoroscopy, interventional radiology and computed tomography, and nuclear medicine imaging both use ionising radiation. Ultrasound and magnetic resonance imaging are described there as types of imaging that do not use X-rays or radioactive substances. That single line explains most of what follows.
The methods side by side
| Method | What it uses | Ionising radiation | Stated duration | Source and date |
|---|---|---|---|---|
| Plain X-ray | Radiation produced by an X-ray machine, passing through the body to create an image | Yes | The X-ray itself takes a few minutes | NHS, reviewed 4 February 2026 |
| Computed tomography | X-rays, in a large ring that moves around part of the body | Yes | Between 10 and 20 minutes | NHS, reviewed 8 November 2023 |
| Magnetic resonance | No X-rays and no radioactive substances; a machine with a short tunnel and a flat bed that moves through it | No | Usually between 15 minutes and 1 hour, but can take longer | NHS, reviewed 11 September 2025 |
| Ultrasound | Sound waves | No | Around 10 to 45 minutes, but it might take longer | NHS, reviewed 25 February 2025 |
| PET | An injected radioactive tracer, given about an hour before | Yes | Around 30 to 60 minutes | NHS, reviewed 4 March 2025 |
| DEXA | Dual energy X-ray absorptiometry, low dose | Yes | Not stated on the NHS page | NHS, reviewed 5 October 2022 |
| Angiography | X-rays with a contrast agent injected into the blood | Yes | Between 30 minutes and 2 hours | NHS, reviewed 30 January 2023 |
Published doses, in millisieverts
Only one official comparison table in millisieverts could be read: Ionising radiation: dose comparisons, published by the UK Health Security Agency on 18 March 2011. Its figures are reproduced here exactly as published, alongside the everyday comparisons the same table carries.
| Item | Dose in mSv |
|---|---|
| Dental X-ray | 0.005 |
| 100 g of Brazil nuts | 0.01 |
| Chest X-ray | 0.014 |
| Transatlantic flight | 0.08 |
| UK annual average radon dose | 1.3 |
| CT scan of the head | 1.4 |
| UK average annual radiation dose | 2.7 |
| CT scan of the chest | 6.6 |
| Average annual radon dose in Cornwall | 6.9 |
| CT scan of the whole spine | 10 |
| Annual exposure limit for nuclear industry employees | 20 |
No other official table of effective dose per examination could be opened. The national diagnostic reference levels published by the UK Health Security Agency on 11 December 2025 are a different quantity altogether: they are given in dose area product, CTDIvol and dose length product, not in millisieverts, and they are described there as generally the third quartile value from a national survey. Treating one as the other is the commonest error in writing about this subject, and it is set out in full in radiation doses.
What this table does not do
It does not rank the methods, and it does not say which one answers any particular clinical question. That decision belongs to the practitioner who justifies an individual exposure under regulation 11 of the Ionising Radiation (Medical Exposure) Regulations 2017, and the way that works is described in IR(ME)R 2017. Nothing here is an assessment of a person, an image or a result.