Citation, DOI, disclosures and article data
At the time the article was created Andrew Murphy had no recorded disclosures.View Andrew Murphy's current disclosures
Noise in computed tomography is an unwanted change in pixel values in an otherwise homogeneous image. Often noise is defined loosely as the grainy appearance on cross-sectional imaging; more often than not, this is quantum mottle.
Noise in CT is measured via the signal to noise ratio (SNR); comparing the level of desired signal (photons) to the level of background noise (pixels deviating from normal). The higher the ratio, the less noise is present in the image.
Noise in a cross-sectional image will equal a decrease in the picture quality and inadvertently will hinder the contrast resolution.
Factors affecting noise
The mAs or the dose of a CT scan has a direct relationship with the number of photons utilized in the examination. A useful relationship to keep in mind is:
2 x mAs = 40% increase SNR
Increasing the dose of the scan will decrease the amount of noise and hence improve the contrast resolution of the image. However it comes at a cost, and balancing the dose with the contrast resolution required for interpretation must be considered when determining examination settings.
Studies that rely on superior contrast resolution will inescapably require a higher dose than examinations that can tolerate a higher amount of noise, for example, liver imaging vs cardiac calcium scores.
The number of photons available to generate an image has a linear relationship to the slice thickness. The thicker the slice, the more photons available; and the more photons available, the better the SNR. However, this is not without a trade-off because increasing the slice thickness will decrease the spatial resolution in the z-axis.
Larger patients will absorb more radiation than smaller ones, meaning fewer photons will reach the detector hence reducing the signal to noise ratio.
Non-linear reconstruction algorithms can cause noise non-uniformity, meaning the intensity of noise varies across the image depending on regional structure. Uniform regions of the image will generally have lower noise levels than highly structured regions.
A variety of metrics are used to measure different qualities of CT noise. Noise has many aspects including magnitude, texture, and nonuniformity.
Noise magnitude is quantified simply by the standard deviation. CT noise magnitude makes up the denominator of the signal to noise ratio.
Noise texture is the visual impression or quality of noise. It can be measured quantitatively by computing the noise power spectrum.
Noise non-uniformity is caused by variation in noise magnitude or texture across the image.
- 1. Euclid Seeram. Computed Tomography. ISBN: 9780323312882