CT extra flashcards
Describe 1st generation CT scanners
Translate rotate system
1. The x-ray beam is picked-up by a single detector.
2. The x-ray source and detector then move together (translate)
3. The two then rotate together to image a different angle
4. This is repeated until a single slice is scanned
5. The two then move down the patient to start imaging a different slice
Describe 2nd generation CT scanners
translate rotate
1. The x-ray beam is picked-up by a row of up to 30 detectors.
2. The x-ray source and detector then move together (translate)
3. The two then rotate together to image a different angle
4. This is repeated until a single slice is scanned
5. The two then move down the patient to start imaging a different slice in the patient
Describe 3rd generation CT scanners
Rotate rotate
1. The x-ray beam hits a row of detectors wide enough to image the whole slice
2. The two then rotate together to image a different angle
3. This is repeated until a single slice is scanned then the array is moved to a different slice (axial scanning). Alternatively, the detector array is continually moved down the patient as it rotates (spiral scanning), This is the most commonly used method today and takes about 0.3 seconds to image a single slice
*High spec scanners use solid state scintillation
**Can be helical
Describe 4th generation CT scanners - what type of detector is used in these?
Rotate fixed
1. There is a fixed complete ring of detectors
2. The x-ray source rotates around to capture a slice
3. Both then move down the patient to begin imaging a different slice
*Uses solid state scintillants - eg cadmium tungstate
How many revolutions per second can be achieved with a modern CT scanner?
2-4 revolutions per second
How much sampling occurs in CT per rotation?
-Continuous beam is sampled approx. 1000 times per rotation
Slice thickness: what determines thickness in single slice and multi-slice CT?
-Single slice CT = determined by collimation. Limited by detector row width.
-Multisclice CT = determined by width of detector rows
What is the equation for calculation of HU? What is a HU?
-CT image: made of pixels along greyscale – level of grey is determined by density of material (also expressed as LAC), which is represented numerically by Hounsfield unit.
-HU: set so that water measures 0 and everything else is relative to this.
HU = 1000 x (μt – μw) / μw
where:
μt = attenuation coefficient of tissue
μw = attenuation coefficient of water
How does iterative reconstruction compare with FPB re streak artefact reduction, noise sensitivity
Interactive reconstruction in CT
-Adjusts image activity content in steps until calculated activity is as close as possible to measured activity
-Reduces streak artefact in comparison with filtered back projection
-Is less sensitive to noise compared to filtered back projection
-Corrects accurately for attenuation-Enables reconstruction of an image, even if projections at some angles are missing.
T/F An edge enhancing convolution kernel will reduce spatial reduction. What is the effect on noise?
-False: edge enhancement increases the differences between neighbouring voxels thereby increasing spatial resolution, this
also increases noise.
True or false: FBP is used to remove the blurring inherent in tomographically acquired data
True
What factors can affect trans-axial resolution in CT?
-Focal spot (smaller the better)
-Detector size: smaller detectors give higher spatial resolution but more detectors mean more partitions (dead space) and reduced overal detection efficiency
-Detector design properties
What factors can affect Z plane resolution in CT?
-Detector slice thickness: wider (in z axis) the detector row, the lower the resolution
-Pitch: data with overlapping slices improves z sensitivity
-Small focal spot is better
What is the HU of air?
(-1000)
What is the HU of lung?
(-300)
What is the HU of fat?
(-90)