NO PEEKING!!

What is this, how is it used, characteristics?

Describe relevant physics associated with image? How is
image made?
Advantages? Disadvantages?
• PHYSIOLOGIC , FUNCTIONAL, + METABOLIC
• PET image = number of decays (counts)
occurring in each voxel
• Simultaneous annihilation photons (2 511 keV)
assumed to occur along line (line of response)
between detectors
• Radiation emitted from radiopharmaceutical
distributed in the patient is registered by external
detectors positioned at different orientations.
• Distribution of radioactivity estimated by -Tumor
identification + localization
• -Resolution = 5 mmcouple with CT, allows for
accurate attenuation correction
• Radionuclides: short-lived, generated by cyclotron,
biologically active positron-emitting nuclides & approx
half-lives:
• 11C (20 min), 13N (10 min), 15O (2 min), 18F (110 min),
62Cu (10 min), 68Ga (70 min), and 82Rb (1 min).
• 18F-labeled fluorodeoxyglucose [18FDG], T1/2 = 1.7 hr
• -Detector: BGO crystal (higher intrinsic efficiency for
511 photons)
• -PMT tube. 2D/3D by moving septa.
• -Use electronic collimation instead of lead collimation,
high sensitivity and efficiency.
You are the new solo physicist in an established
clinic. Shockingly, they treat all of their H&N
cases to a GTV with no margin. You are
outraged! What do you do?
• What margins do you need? What do they
depend on?
• How can you determine the margin?
• Define an equation, confidence level, and
reference material available
Internal Margin (IM): Variations in size, shape, and position of the CTV
relative to anatomic reference points; e.g., filling of bladder,
movements of respiration. The internal variations are physiological
ones, and result in change in site, size, and shape of the CTV
• Internal Target Volume (ITV): Volume encompassing the CTV and
IM. (ITV = CTV + IM)
• Set-up Margin (SM): Uncertainties in patient positioning and
alignment of therapeutic beams during treatment planning, and
through all treatment sessions. The uncertainities may vary with
selection of beam geometries, and may depend on variations in
patient positioning, mechanical uncertainities of the equipment (e.g.
sagging of gantry, collimators, or couch), dosimetric uncertainities,
transfer setup errors from CT simulator to treatment unit, and
human factors. These may vary from center to center, and from
machine to machine
• Planning Target Volume: PTV = CTV + IM + SM. The penumbra of
the beam(s) is not considered when delineating the PTV. However,
when selecting beam sizes, the width of the penumbra has to be
taken into account and the beam size adjusted accordingly
• PTV includes: RANDOM & SYSTEMATIC ERRORS
 RANDOM: respiration, pt movement, bladder,
bowel, daily setup
SYSTEMATIC: weight loss, technical errors,
tumor reduction
• Simple margin recipe: van Herk
How to get Random & Systematic?

Study IGRT trends in previous tx with similar

immobilization devices. Calculate population
statistics (mean, stdev). Apply to vanHerk.
 Define each image type.
How do we generate these images?
Name nearby structures, dose limiting structures.
What are we treating? Dose?
Adv/disadv?
What is this? Frequency of use?
Process for analyzing? Specs?
Can we use it for Tomo MVCT?
Define HU equation, air, water,
lung?
 Rod Material Electron Density Physical Density g/cm3
Relative to Water
Lung (LN-300) 0.28 0.30
Lung (LN-450 0.40 0.45
Adipose (AP6) 0.90 0.92
Breast 0.96 0.99
CT Solid Water 0.99 1.02
Brain 1.05 1.05
Liver (LV1) 1.07 1.08
Inner Bone 1.09 1.12
Bone (B200) 1.11 1.15
Bone (CB2-30% Mineral) 1.28 1.34
Bone (CB2-50% Mineral) 1.47 1.56
Cortical Bone (SB3) 1.69 1.82
True Water 1.00 1.00
Optional Titanium Insert 3.79 4.59
• ED to CT number HU to RED conversion, 120 kVp

1500
conversion: 1000
-commissioning 500

HU
-annually 0
HFH CT1
HFH CT2
WBC

-calibration/major -500 DCO

Macomb

change (BASELINE,
System Average
-1000
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

TPS) Relative Electron Density

• Check water
uniformity daily
 A C
B

• What image types? Advantages/disadv

• Define A, B, C. Define patient orientation
• What is arrow region? Describe physics.
• CT & MR, spleen/liver
• Contrast agent: higher Z (barium, Iodine)
What is this and two examples of how it is
used?
What QA/Frequency required?
Imaging components? QA?
Simulators:
1. Reproduce geometric conditions of radiation therapy equipment so
undergo same mechanical checks as accelerators.
2. Checked for image quality according to established guidelines
Define DQE.
Describe the
graph in detail
Typical values?
• DQE or Detective Quantum Efficiency
measure of combined effect of noise and
contrast performance of an imaging
system,
• Expressed as a function of object detail, or
spatial frequency.
• Signal-to-noise degradation due to the
imaging system to photon statistics.
• DQE = SNR^2 (out) / SNR^2 (in)
• WANT DQE = as close to 1 as possible.
• FLAT PANEL DIGITAL DET = 65%. Computed
radiography = 35%. Screen film = 25% (slower,
require more exposure for image).
• Higher DQE, flat panel detectors use dose more
efficiently, can provide lower noise images over
range of spatial frequencies
• FLAT PANELS HIGHER DUE TO: better
quantum absorption and conversion efficiency
 A

• Identify
components
YAXIS, UNITS?

• Types, uses in
RT?
• What is latitude?
Speed?
B
C • Define contrast.
• What is net OD?
X-Axis? Units?
• ODnet = OD − OD(base + fog) OD(base + fog) LIMIT < 0.2, initial darkening w/out
exposure to radiation due to:
• Base = absorption of some light in film base (tinting), ~0.07
• Fog = developed Ag grains that appear even in the absence of radiation exposure
(from background radiation). Typically 0.05 for fresh film, increases over time
• Optical density: The degree of blackening of radiograph. The OD is defined
as OD = log10 (I0/I) and is a function of dose. I0 is the initial light intensity
and I is the intensity transmitted through the film, practically, OD = 0 – 4.
Transmittance = I/I0.
• Exposure should occur in straight line portion of curve. Average slope of
curve = average gradient.
• Films w/ higher avg gradient  higher contrast
• Max value of slope = gamma (slope @ inflection point of
• curve)
• Dynamic range/latitude:
• The exposure range over which acceptable optical densities are produced
(linear portion of H&D curve)
• Speed:
• Film speed (sensitivity) = position of the characteristic curve at a specific
value of the exposure axis.
• Film speed = 1 / Exposure in R required for an OD of 1.0 above base density
• SLOWER SPEED FILM: toe of H&D curve longer, smaller granules,
fewer x-rays to form image
• FASTER: toe of H&D curve shorter, larger granules, noisier image
• Define grid, grid ratio.
• Why do we need it? Materials?
• Is it used in conventional sim? Fluoro
setup? EPID? Hard films?
• GRID = dense, highZ material (Bucky), lead,
~0.05 mm thick, interspace = fiber/plastic/or
aluminum. Al = further reduce scatter & are
sturdier
• GRID Ratio = height of strips / distance between
strips (typically use 8-12)
• ALL GRIDS IMPROVE CONTRAST
• GRID NOT AS EFFECTIVE FOR HIGHER V
• Pt exposure increases w/grid ratio
• Used in diagnostic departments
(film/screen/grid combinations)
• CBCT can use a scatter rejection grid =
10:1
• FLUORO: Uses antiscatter grid
Describe what you
see.
What are the 2
images?
Algorithms?
Resolution?
Pitfalls of doing this?
Name relevant
landmarks to
evaluate match
• Image fusion– optimization,
transformation, similarity matrix.
Optimizer: seeks global max of cost value.
Iterative modifying params. @each step,
parameters used by cost function to calculate a
new cost value/function

Similarity matrix: how similar two images sets

are. Mutual info is one of evaluation way and it
supports multi-image modality.

MI: measures amt of information one variable

contains about another variable. Samples
data, computes joint histogram. For well
registered, joint histogram = peaked = high MI.
 Clinical process
• Rigid registration w/linear optimization.
Rotates/translates. Re-evaluates MI in 2
images.
• Can use VOI/contour of interest taken into acct
• Then do intensity based matching
• Verify fusion after autoregistration
• Fiducials, landmarks