# week 4, lec 1- image creation Flashcards

1
Q

what is the image formation process?

A

CT gantry, patient and table, attenuation measurements and digital date –> CT computer –> CT numbers –> CT console –> PACS

2
Q

what are the sequence of events after signals leave the detectors

A
• CT detectors 
• Pre-processing 
• Reformatted raw data  - Convolution with filter
• Image recon algorithm
• Reconstructed images of CT numbers 
• Image storage, display etc.
3
Q

what is data acquisition

A

Taking ‘real life slices’ and converting them to numeric values

4
Q

what is attenuation

A

reduction in intensity of x-ray beam as it passes through an object (the patient)

The remaining x-rays are transmitted and reach the detectors

5
Q

what is the linear attenuation coefficient

A

Describes how easily a volume of material can be penetrated by a beam

e.g. the fraction of a beam of x-rays or gamma rays that is absorbed or scattered

6
Q

what is the linear attenuation coefficient dependent on?

A

the x-ray beam average energy (Eave)

• Change Eave and µ changes
• To get a true measure of µ, the x-ray beam must be monochromatic (1 energy for all photons)
7
Q

what is fourier transform

A
• A mathematical function that converts a signal in the spatial domain to a signal in the frequency domain
8
Q

what is the spatial domain

A

Spatial domain, we directly deal with the image matrix.

• X-ray beam passes through patient, image profile denoted by f(x) obtained
• Use of these coefficients allows assists in reconstruction of CT image
9
Q

what is the frequency domain

A

-In frequency domain, we deal with the rate at which the pixel values are changing in spatial domain

10
Q

what is the reconstruction problem in CT

A

There is an object of unknown x-ray attenuation characteristics

• Beam is NOT homogenous
• Need to determine the individual linear attenuation coefficient (µ) for each voxel
• Attenuated beam exits the subject it comes into contact with a series of electronic detectors
• Raw data needs to be converted into a numeric map
• X-ray tube and detectors rotate through 360 degrees
• At each angle (typically <1 degrees, so more than 360 angles)
• The x-ray beam passes through the patient and is attenuated by the patient’s body
11
Q

provide an overview of the principle behind CT

A
• Data is collected by the detectors (forms the slice profile)
• The x-ray beam passes through the patient and is attenuated by the patient’s body
• Intensity values are converted to digital values
• Slice profiles typically are “filtered”
• Detectors gather attenuated x-rays which are then amplified and converted into digital signal as RAW data (aka projections)
12
Q

list 5 image reconstruction approaches

A
• Simultaneous linear equations
• Back projection
• Filtered back projection
• Fourier reconstruction
• Iterative techniques (now-days)
13
Q

what are 3 assumptions of heterogenous beam attenuation

A
• Attenuation is not exponential
• Quality and quantity of photons change
• Lower energy photons absorbed, higher energy transmitted (photons that interact with detector)
14
Q

what is the lambert-beer law

A

Describes what happens to the attenuation of the original intensity of photons

15
Q

what is the objective of CT

A

To calculate the linear attenuation coefficient

This indicates the amount of attenuation that has occurred (between primary beam and beam that interacts with detectors)

16
Q

what are CT numbers/ Hounsfield Units

A
• CT numbers or Hounsfield units are used instead of linear attenuation coefficients (because thy are solid numbers; no decimals)
• CT number = Digital image
• CT number is a number related to the linear attenuation co-efficient of the object
17
Q

what is the scan profile

A

A measure of the intensity (I) by each detector at a given angle of rotation

18
Q

what is the reconstruction process

A

converts scan profile data into a matrix of µ’s, then change the µ to CT no’s / HU. Reconstruction is a mathematical process using a variety of methods

19
Q

what is the process of filtered back projection

A
• Changing the filter yields a trade-off between noise and sharpness of the image
• It is an analytic reconstruction algorithm designed to overcome the limitations of conventional back-projection; it applies a convolution filter to remove blurring
• It utilizes simultaneous equations of ray sums taken at differing angles of a sine wave to compute the values of attenuation coefficients
20
Q

why is filtering done on the scan profiles

A
• Process is linear: if it works for a point object (at any location), then it will work for any image
• results in a sharper reconstructed image
21
Q

what is Iterative reconstruction

A
• IR does offer this ability to have high spatial and high contrast resolution and can do so at a lower dose
• Iterative reconstruction is an alternative reconstruction algorithm that can remove the noise from low-dose images using a variety of mathematic models
• Reduces rad exposure while improving image quality
22
Q

how does iterative reconstruction work

A
• ASIR uses matrix algebra to continually process and compared to assumed ideal models over and over to improve the image voxel-by-voxel until they converge to a final value
• The software builds an image and then revises it with scores of reiterations to enhance image quality
• Radiation exposure from multi detector is high
• The data set is processed in a continuous loop where calculations are performed to create the different images
23
Q

A
• IR software can help reduce streaking, metal and blooming artefacts
• Increase the contrast-to-noise ratio in bariatric imaging (higher radiation doses) by better filtering noise
• Increase iodine opacification to help reduce contrast dosage
24
Q

assumptions about reconstruction from scan profiles in helical/ spiral CT

A
• Additional maths is needed for fan beam reconstruction.
• Helical (or spiral) CT has the patient move while the tube and detectors are rotating
• Pitch is a parameter used to describe the “stretch” in the Z direction of the helical path
25
Q

what is helical reconstruction

A
• Will still use 1 rotation of data in the image reconstruction
• 1 scan profile data will be at the exact location of the axial recon plane, other data will be a various distance from that plane
• The closer the data is to the recon plane, the greater its importance in the image
• The further the scan data is from the recon plane the less importance or weighting it will have
• This is a process called interpolation.
26
Q

what is interpolation

A

-Interpolation is a method of calculating new data points within the range of known or existing data points