Dual-energy CT, also known as spectral CT, is a computed tomography technique that uses two separate x-ray photon energy spectra, allowing the interrogation of materials that have different attenuation properties at different energies.
Technique
Whereas conventional single-energy CT produces a single image set, dual-energy data (attenuation values at two energy spectra) can be used to reconstruct numerous image types:
weighted average images (simulating single energy spectra)
virtual monoenergetic images (attenuation at a single photon energy rather than a spectrum)
-
material decomposition images (mapping or removing substances of known attenuation characteristics, such as iodine, calcium, or uric acid)
virtual non-contrast images (iodine removed)
iodine concentration (iodine maps)
calcium suppression (calcium removed)
uric acid suppression (uric acid removed)
electron density maps
effective atomic number (Zeff) maps
Physical principles
X-ray photons primarily interact with matter via the photoelectric effect, and Compton scattering producing the diagnostic images used in medicine today.
When an atom undergoes the photoelectric effect, the electron from that respective K-shell otherwise referred to as the inner shell, is ejected via the incident photon. As that electron is excited, vacant space is 'filled' by a neighboring electron, releasing energy as a photoelectron.
In short, when a photon has sufficient energy to overcome the electron's binding energy in the K-shell, that atom undergoes the photoelectric effect.
Each substance owns a unique K-shell binding energy; known as the K-edge. There is a significant spike in attenuation that results just beyond the energy of the K-edge, this peak is unique to every material and holds valuable information about the substance's composition.
The different photoelectric energies and K-edges are the bread and butter of dual-energy CT. Although most elements in the human body have very low K-edges (0.01-0.53 keV), elements like iodine and calcium have higher K-edges of 33.2 keV and 4.0 keV respectively, making them sufficiently larger than surrounding structures and are particularly important in the clinical setting 1-3.
For instance, at 80 kVp a structure that contains no (introduced) iodine, such as the liver, has an attenuation based on its K-edge of x, yet when iodine (33.2 keV) is introduced into that same structure, it has a higher attenuation of y bringing it closer to 80 kVp.
As 80 kVp is closer to 33.2 keV than 140 kVp, the structures containing iodine will retain less attenuation as the kVp progresses beyond the K-edge of iodine. Therefore, when using two energies, it is possible to delineate structures based solely on their attenuation differences between 80 kVp and 140 kVp.
A dual x-ray source, tube A (140 kVp) and tube B (80 kVp or 100 kVp) with an angular offset of 90 degrees are preferred offsets for a dual source scanner in the current literature 1-5.
Image acquisition
There are different DECT acquisition technologies available from different vendors. These can be broadly classified as techniques that occur before the patient is scanned (prospective) which need to be pre-selected and those that occur after the patient is scanned (retrospective) which do not need to be pre-selected.
Prospective techniques
-
dual-source
two x-ray tubes producing different voltages (kVp) offset at approximately 90°
reconstructed in the image space
limited field of view (FOV) as both detectors can not be the same size
excellent temporal resolution as both datasets were acquired at the same time
-
single-source consecutive
two helical scans are consecutively acquired at different tube potentials followed by coregistration for post-processing
reconstructed in the image space
full field of view
poor temporal resolution as the patient is scanned twice (therefore increased dose)
-
single-source twin-beam
two-material filter splits the x-ray beam into high-energy and low-energy spectra on the z-axis before it reaches the patient
-
single-source sequential ("rotate-rotate")
each x-ray tube rotation is performed at high- and low-tube potential
reconstructed in the image space
full field of view
poor temporal resolution as the patient is scanned twice (therefore increased dose)
-
single-source rapid kilovoltage switching (fast kVp-switch)
the x-ray tube switches between high- and low-tube potential multiple times within the same rotation
reconstructed in the projection space
full field of view
slight reduction in temporal resolution due to tube rotation
Retrospective techniques
-
dual-layer DECT ("sandwich")
the top (innermost) layer of the detector absorbs low-energy photons while high-energy photons pass through to the bottom (outermost) layer
reconstructed in the projection space
full field of view
excellent temporal resolution as both datasets were acquired at the same time