2. INTRODUCTION
• Since its invention in 1971, technologic advances in computed
tomography (CT) have expanded its application.
• Advances in hardware as well as in image reconstruction and
processing have enabled faster scanning at lower radiation dose for
contemporary multidetector-row CT (MDCT) scanners.
• The most notable software advances have come in the form of dual-
energy image processing, iterative reconstruction, and automatic kV
selection techniques, which have proven to be key in several
procedures.
3. • Hardware advances in CT technology have resulted in the introduction
of several new MDCT scanners with capabilities of i)greater x-ray tube
power,
ii)faster gantry rotation times
iii)wide-area and dose-efficient detector array systems
iv)dual-energy capabilities with single-source, twin-beam, and
sandwich layer detector systems.
4. 1.CT FLUOROSCOPY
• Combines the conventional advantages of both CT and fluoroscopy
• Has an important role in image-guided interventions where real-time
imaging is required.
• CT fluoroscopy combines the cross-sectional image targeting provided
by CT with the real-time imaging, tracking and movement perception
of fluoroscopy for interventional procedures
• It allows continuous update of images at a fixed position and is
commonly used for CT-guided biopsies and fluid drainages
5. TECHNICAL CONSIDERATIONS
• Video monitor will need to be displayed in the scanning room
• An operator panel is required in the scanning room – with controls available for table
movement, gantry lift, laser light control and fluoroscopic factors. Exposures will
usually be activated using a footswitch
• Involves an x-ray tube current of 30-50 mA, compared with conventional fluoroscopy
with approximately 4 mA, or conventional CT with approximately 150-400 mA
• Need for additional beam filtration to decrease patient radiation exposure
• Consideration for radiation exposure to the personnel in the room
• Multislice machines have finer z-axis resolution, which improves localization accuracy
• CT fluoroscopy requires special techniques for image reconstruction, due to the need
for rapid imaging feedback
6. CT Fluoroscopy reconstruction
• CT fluoroscopy uses a partial reconstruction technique, which means
that data from the last 360 degree set is reconstructed and displayed
after every N° (30°/45°/60°) of tube motion.
• A frame rate of 6-12 frames per second is generally used.
• The delay between each image displayed depends on frame rate and N°.
• For example, for 60 degree updates and 6 frames per second, the
image is updated every 0.17 seconds (compared with conventional
fluoroscopy which is around 25 frames per second).
• Because of this, a relatively high level of noise will be apparent in soft
tissues.
8. To patient…
• The patient surface dosage may range between 2 and 10 mGy/sec,
with exposure times lasting up to 200 seconds.
• There is potential for radiation-induced deterministic effects to the
skin in long, high-dose CT fluoroscopic procedures.
• For safety reasons, departments should monitor the patient dose
based on departmental phantom studies and cease the procedure if
the deterministic threshold is reached.
9. To operator…
• The most significant area of radiation exposure to the radiologist is the hands.
• The absorbed dose to the hands in the direct beam is approximately 1.1 mGy
per second.
• Other at risk organs from a radiation exposure perspective include the
thyroid and lens of the eye.
• Departments will usually have a radiation safety threshold for staff on a per
annum basis.
• Deterministic and stochastic effects could be developed if the radiation
exposure exceeds 500 millisieverts per year and 50 mSv per year, respectively.
10. Dose reduction strategies
• Minimize CT fluoroscopic scanning time. Consider incremental needle
insertion outside the gantry, with intermittent short CT fluoroscopic
‘flashes’ to check for needle tip location
• Use of lower-exposure CT fluoroscopic techniques (use lower mAs)
• Reduce section thickness (e.g. use 2 mm or 5 mm thickness, rather
than 10 mm)
• Distance
• keep fingers out of the primary beam
• maximize distance from the primary beam and patient (inverse square law)
11. Cont…
• Shielding and other protective devices
• Use purpose-built needle holders to increase the distance of the hands from the
primary beam.
• Use a lead or tungsten-antimony drape placed over the patient, adjacent to the scan
plane.
• A lead apron placed 2.5 cm caudal to the scan plan reduces exposure by
approximately 70%
• use of a vertically fenestrated gantry shield/drape can reduce dose by up to 30%
• use of personal lead protection e.g. lead glasses (greater than 0.1 mm lead reduces
eye dose by 50%), lead aprons, thyroid shields (reduce dose by > 95%), lead gloves
• use the distal side of the gantry, where shielding can be provided by electronic racks
inside the gantry
12. 2. CT COLONOSCOPY
• Also called virtual colonoscopy or pneumocolon
• Mainly indicated for colorectal cancer, assessment of strictures,
evaluation of obstructions
13. Technique
• The colon should be clean and completely distended
• Colonic distention is critical to technical success for proper
intraluminal evaluation of the large bowel and can be achieved via a
pressure-regulated device.
• Patient given an antispasmodic agentIV/IM hyoscine-N-butylbromide
(Buscopan), an antimuscarinic drug reduces colonic motion, leading
to higher quality images with reduced patient discomfort
• IV glucagon is used in some countries/institutions as a first or second-
line antiperistaltic
14. contraindications
• acute inflammatory conditions such as acute diverticulitis, active
stage of ulcerative colitis or Crohn disease
• recent abdominal or pelvic surgery
• CTC cannot be performed if a colostomy is present as there is no
natural sphincter mechanism to retain the tube
• general CT contraindications e.g. pregnancy, claustrophobia, etc.
• history of severe allergy / anaphylaxis to iodinated contrast media
• patients at high risk for a gastrointestinal tumor (e.g. Lynch syndrome)
may not be good candidates for CTC screening
15. Data acquisition
• CT scanning is ideally performed on a multi-detector computed
tomography (MDCT) scanner in both supine and prone positions with
a thin collimation
• image review with the use of two-dimensional (2D) and three-
dimensional (3D) displays is strongly advised for optimal evaluation
16. Advantages
• Virtual colonoscopy has several advantages over conventional
colonoscopy:
• less invasive procedure, therefore complication rate lower
• takes less time
• can visualize colon beyond the obstruction or narrowing
• detects extracolonic pathology
17. Disadvantages
• residual fecal material can give rise to wrong interpretation
• biopsy specimen cannot be taken at the time of the procedure
• exposure to ionizing radiation
19. SPECT
•SINGLE PHOTON Unlike PET, which uses dual annihilation
photons for image creation,SPECT uses single photons from gamma decay.
•EMISSION emission imaging that is different from transmission
imaging like X-ray or Reflection-based imaging like Ultrasound)
•COMPUTED uses algorithms (NOT Geometric Tomography)
•TOMOGRAPHY produces tissue slices (NOT planar imaging)
20. WHAT IS IT?
•Nuclear Medicine imaging modality that combines conventional
scintigraphic and computed tomographic methods.
•which involves the use of radionuclides injected
intravenously into the body,
•to produce a 3D distribution of the gamma rays emitted by the
radionuclide,
•giving physiological information about the organ of interest.
21. • SPECT combines conventional scintigraphic and computed
tomographic methods.
• It gives a 3D functional image rather than a 2D planar image that is
given by planar scintigraphy
• It restricts superimposition of active and nom active functional layers.
• It has a computer that collects the information emitted by the gamma
rays and translates them into two-dimensional cross-sections.
• These cross-sections can be added back together to form a 3D image
of your brain
22. • The radioisotopes typically used in SPECT to label tracers are iodine-
123, technetium-99m, xenon-133, thallium-201, and fluorine-18.
• These radioactive forms of natural elements will pass safely through
your body and be detected by the scanner.
• Various drugs and other chemicals can be labeled with these isotopes.
23. DESIGN
• SPECT machines combine an array of gamma cameras (ranging from
one to four cameras) which rotate around the patient on a gantry.
• SPECT may be also combined with a separate CT machine in a form
of hybrid imaging; single photon emission computed tomography-
computerized tomography (SPECT-CT) mainly for the purposes of
attenuation correction and anatomical localization .
25. PRINCIPLE
• Gamma cameras rotate around the patient providing spatial
information on the distribution of the radionuclide within tissues.
• The use of multiple gamma cameras increases detector efficiency and
spatial resolution.
• The projection data obtained from the cameras are then
reconstructed into three-dimensional images usually in axial slices.
• When SPECT-CT is used, attenuation correction and higher resolution
anatomical localization can be achieved 1
.
26. • Images may be taken in 2 modes:
• Continuous Acquisition: while camera heads are in motion
• Step and Shoot: camera heads stop at defined angles to acquire image.
•Rotation may be full 360o around the body, or 180o with projection algorithms used
to construct for the other half. 180o is most common generally, 360o is common for
brain SPECT
27. Image reconstruction
• After acquiring a 2D projection image , axial position
reformating occcurs
• The reformatted projection is represented as a3D function.
• This forms a sinogram. This is important as it detects
artefacts.
• After a series of projections there might be overlap of
anatomy
• Conventionally there was blurring but with spect there is
use of computer algorithms to remove the overlying
structures completely, and solves blurring effect.
30. • Images are usually in transverse plane but pixels can be reordered to
produce coronal and sagittal image planes.
• Fourier transforms and mathemartical filtering of data is done to
remove noise in data to remain with anoise free final image.
31. Image reconstruction
• The goal of image reconstruction algorithms is to calculate
accurately the 3D radioactive distribution from the acquired
projections
• The reconstruction of tomographic images is made by two methods:
1. Iterative method
2. Filtered Back Projection
32. Iterative
• Iterative reconstruction starts with an initial estimate of the image.
• Then a set of projection data is estimated from the initial estimate
using a mathematical process called forward projection.
• The resulting projections are compared with the recorded
projections and the differences between the two are used to update
the estimated image.
33. • The iterative process is repeated until the differences between the
calculated and measured data are smaller than a specified preselected
value.
The iterative reconstruction methods include:
1. algebraic methods like:Algebraic Reconstruction Technique (ART)
2. statistical algorithms like:
• Maximum Likelihood Expectation Maximization (MLEM), or
• Ordered-Subsets Expectation Maximization (OSEM)
34. • Iterative image reconstruction methods allow the incorporation of
more accurate imaging models rather than the Radon model assumed
in the FBP algorithm
• These include scatter and attenuation corrections, as well as
collimator and distance response, and more realistic statistical noise
models.
36. Filtered back projection.
• Back projection technique redistributes the number of counts at each
particular point back along a line from which they were originally
detected.
• This process is repeated for all pixels and all angles. The limited
number of projection sets has as a result the creation of a star
artifact and the blurring of the image.
• To eliminate this problem, the projections are filtered before being
back projected onto the image matrix.
37. • The filters used in FBP are simply mathematical equations that vary
with frequency.
• They attempt to achieve different purposes, such as
1. Star artifact reduction,
2. Noise suppression,
3. Signal enhancement and restoration
38. Differences of FBP and Iterative
FBP ITERATIVE
FAST RELATIVELY SLOW
Produce less accurate images of radioactive
distribution.
Produce accurate images of radioactive distribution.
Less sensitive More sensitive
40. 4.PET
• Positron-emission tomography (PET) is a nuclear medicine functional
imaging technique that is used to observe metabolic processes in the body
as an aid to the diagnosis of disease.
• In PET, organic molecules labeled with positron-emitting radionuclides are
injected or inhaled, and the high-energy photons produced by annihilation
events are detected by paired, integrated crystal detectors.
• The system detects pairs of gamma rays emitted indirectly by a positron-
emitting radionuclide (tracer), which is introduced into the body on a
biologically active molecule.
• Three-dimensional images of tracer concentration within the body are then
constructed by computer analysis.
41. • In modern PET-CT scanners, three-dimensional imaging is often
accomplished with the aid of a CT X-ray scan performed on the
patient during the same session, in the same machine.
• Other positron-emitting radionuclides that are also used are carbon
11, oxygen 15, nitrogen 13, fluorine 18, and rubidium 82.
42. • The positron emitting isotope administered to the patient
undergoes β+
decay in the body
• The annihilation reaction results in the formation of two high energy
photons which travel in diametrically opposite directions.
• Each photon has an energy of 511 keV.
• Two detectors at opposite ends facing each other detect these two
photons traveling in opposite directions, and the radioactivity is
localized somewhere along a line between the two detectors.
• This is referred to as the line of response.
43. Artefacts
• When PET is combined with CT, the CT imaging can be used to generate an
attenuation map which is used to correct the PET imaging for attenuation.
This attenuation correction can add a number of further artifacts.
• Breathing artifacts causes the 'mushroom effect' where an artifact is
sometimes seen in the lung bases because of the different phases of
respiratory motion
• Metallic implants and prostheses such as joint prostheses can create
significant artifact on PET images as the attenuation correction cannot deal
with/correct for markedly high densities
• Truncation -Occurs when CT field of view is limited and PET field of view is
usually larger.
44. Summary
• This is a functional modality that has anatomical correlation
algorithms.
• Anatomical correlations show pathological sites of accumulation and
can differentiate them from normal physiological uptake.
• Both are housed in the same gantry.
• Limited in that it causes confusion of the precise location of the
photons
45. CT Angiography
• It is a modality that produces detailed images of both blood vessels
and tissues in various parts of the body after contrast media has been
injected.
• Uses include examining blood vessels and the organs supplied by
them in various body parts, including: brain ,neck , heart , chest
abdomen (such as the kidneys and liver) ,pelvis ,legs and feet ,arms.
46. Injection techniques.
1. Test bolus
- in this method , a target location is selected from the scanogram then a small test
bolus is injected intravenously.
-Then acquire a low dose dynamic scan at specified location during injection
2.Bolus tracking
- there is need to select the trigger location then acquire a reference image
- place region of interest in vascular structure of interest then inject contrast scan
- acquire low dose dynamic scans while monitoring attenuation in the region of interest.
3.Time/fixed delay
- in this method a fixed delay time is set according to the region of interest and thus
scanning starts just after the delay time
48. Intelligence
• Ability to perceive information and to retain it as
knowledge to be applied towards adaptive
behaviours.
49. AI
• Is intelligence demonstrated by machines.
• We deem a computer to exhibit artificial intelligence when it
performs a task that would normally require intelligent action by a
human.
• Any device that perceives it's environment and takes action that
maximise it's chance of successfully achieving its goals.
• Often used to describe machines that mimic cognitive functions that
humans associate with the human mind.
51. • Learning- simplest is learning by trial and error
• Reasoning- is to draw inferences appropriate to the situation.
• Problem solving- particularly in AI, may be characterised as a
systematic search through a range of possible actions in order to
reach some solution.
52. Cont
• Since the development of the digital computer in the 1940s, it has
been demostrated that computers can be programmed to carry out
very complex tasks.
• Some programs have attained the performance levels of human
experts and professionals in performing certain specific task.
53. • Machine learning- is an application of AI that provides system the
ability to automatically learn and improve from experience without
being explicitly programmed
• It is a branch of AI based on the idea that systems can learn from
data, identifying patterns and make decisions with minimal human
intervention.
55. • Deep learning- is an AI function that mimics the workings of the
human brain in processing data, for use in detecting objects,
recognising speech and translating languages.
• AI algorithms particularly deep learning have demonstrated
remarkable progress in image-recognition tasks.
56. • Historically in radiology practice, trained physicians visually assessed
medical images for the detection, characterisation and monitoring of
diseases.
• AI methods excel at automatically recognising complex patterns in
imaging data and providing quantitative assessments of radiographic
characteristics.
57. Convolutional Neural Network
In deep learning a CNN is a class of deep neural networks most
commonly applied to analyzing visual imagery.
Has applications in medical image analysis.
Convolutional networks were inspired by biological processes in that
the connectivity pattern between neurons resembles the organization
of the animal visual cortex.
58. • With the growing use of CT scanning, it contributes to 62% of the
radiation dosage that people incur from all imaging modalities
• A team of bioengineers developed an AI technique that uses image
post- processing to rapidly convert low dose CT scans to images of
superior quality compared to low dose scans that don't use AI
technique.
• This deep learning image reconstruction technique intergrates low
radiation dose CT images with emerging neural network methods
and offers comparable images at much higher speed as those
produced with iterative reconstruction methods.
59. Modular Neural Network
Artificial neural network characterized by a series of independent
neural networks moderated by some intermediary
The modularized neural network for CT image reconstruction
progressively reduces data noise
The noise causes decreased image quality as a result of low radiation
dose CT scanning.
60. • The problem according to research is that those algorithms to do with
iterative reconstruction methods sometimes remove useful
information or falsely alter the image.
• So researchers found that their deep learning method is also much
quicker and allows the radiologist to fine-tune the images according
to clinical requirements.
61. COVID-19
• Chest CT scans are a useful tool for evaluating and diagnosis
symptomatic patient suspected to have the virus.
• Researchers leverage patient's clinical information, including blood
test results showing any abnormalities in white blood cell count
• The group then integrated data from those CT scans with the clinical
information to develop an AI algorithm, which copies the work flow a
physician uses to diagnose COVID-19 and provides a positive or
negative diagnosis prediction researchers noted.
62. • Additionally the algorithm recognise 68% of COVID-19 positive cases
when radiologist found these negative due to negative CT
appearance
