Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Imaging Review CCPM > Mike's notes > Flashcards

Mike's notes Flashcards

1
Q

Planck’s constant

A

6.626*10^-34 Js

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

EM spectrum from low frequency to high

A

radio, microwave, infrared, UV, xrays and gamma rays
-goes from 10^5 to 10^21 Hz, 10^-9 to 10^6 eV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

how to reduce heel effect

A

increase anode angle, increase SID, or decrease field size

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what energy photos does inherent xray tube housing stop

A

< 15 keV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

for what energies is compton dominant in water?

A

25 keV to 25 MeV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

what happens if SID incorrect when using grid?

A

Scatter grid would not be aligned with the diverging beam thus less effective (also removes primary),

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

why is patient as close as possible to detector?

A

reduce focal spot blurring
magnification increases focal spot blurring

however it is not eliminated (patient non-zero thickness)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

what type of scintillator is CR

A

BaFBr

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

By how much do PMTs multiply charge

A

Photocathode:emit e-per every 5 light photons
Dynodes:100V increments, eject 5 e-for each e-striking
If 10 dynodes ⇒amplification= 510≈ 10 million

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

issue with CCD

A

poor optical lens coupling efficiency = quantum sink

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

typical conversion efficiency of common screens

A

10%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

why use rare earth screen

A

higher absorption efficiency, thus faster

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

energy of a light photon in screen film

A

~ 2 eV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

The use of gonadal and fetal shielding in diagnostic x-ray is discontinued (may cause ­ dose if small part within FOV due to 1. repeated exam, 2. tube output ­ by automatic exposure control ). See: AAPM 2019 position statement in references

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

what % of W target radiation is characteristic?

A

5% at 80 kV, 10% at 100 kV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

what is a mA

A

6.24*10^15 electrons/s

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

In chest radiography, patient’s lung diaphragm should be placed at anode or cathode side?

A

Diaphragm on cathode side.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is the difference between quantum detection efficiency and energy absorption efficiency? Which one is more appropriate for radiography imaging systems?

A

Quantum detection efficiency is the fraction of photons detected (interacted). Energy absorption efficiency is the fraction of energy absorbed by the detector thus it excludes part of the energy leaving the detector after photon interaction (e.g. characteristic x-ray). The later is more important as most detectors are energy integrators not photon counters.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

quantum sink in CCD

A

The size of CCD is limited due to the dimensions of crystalline silicon wafers.
For an ideal detector, quantum sink should only happen at the first stage of x-ray absorption, making the detector x-ray quantum limited. However, in the system in the figure above, secondary quantum sink happens due to loss of light photons.
(poor optical lens coupling efficiency)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

ODs of two superimposed films are additive (why?)

A

Is additive since log convers multiplication to addition.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

speed film

A

High speed films has a smaller toe (and slow films larger toe).
Slower film requires more dose to the patient (why?).
Film speed = 1 / (Exposure in R required for an OD of 1.0 above base + fog)
The film speed mainly depends on the size of silver bromide grains. A film with larger grains is faster than a film with smaller grains.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

typical entrance patient dose in fluoro

A

20 mGy/min

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

W target in mammo

A

no characteristic xrays because kedge is around 59 keV which is outside mammo kV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

how does using air gap instead of grid in mammo affect dose?

A
  1. reduced by a factor of 2-3 since grid is removed
  2. increased by the same factor (2-3 x) due to inverse square law.
    Thus, the overall dose is comparable to the grid technique. However, since FOV ¯, smaller part of the breast receives this dose.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

dose from tomo mammo

A

similar to conventional ammo (2 mGy)
tomo: acquire 11-51 images over limited 15 degree angle

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

ideal MTF

A

1 for all frequencies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

what is xray transmission through patient modelled by?

A

binomial process

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Probability of event B, given that event A has occurred

A

Pr(B|A)=Pr(A and B)/Pr(A)
if A and B are independent, Pr(A|B)=Pr(B)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

EXAMPLE of <a> given P(a)</a>

A

e.g. for rolling a six-sided dice, ai=1,2,…6 all having same probability 1/6. Thus, <a> = 1x1/6 + 2x1/6 +…+ 6x1/6 = 3.5</a>

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

noise power spectrum

A

total noise in power in a(x) at each frequency
-measure of the variance per unit spatial frequency
total power in signal = integral from minus to plus infinity of f(x)^2dx OR F(u)^2du

aka wiener spectrum and covariance spectrum

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

MTF relation to LSF

A

MTF = FT(LSF)
For digital systems MTF ~ sing(piFd) due to squares- since in fourier space

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

equation for NEQ

A

(snrOUT)^2=neq=(MTF/NPS)^2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

equation for DQE

A

DQE = NEQ/(SNRin^2)=NEQ/qbar = (SNRout/SNRin)^2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

variance of d(x) in terms of NPS

A

variance = integral from minus infinity to infinity of NPS(u)du

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

NEQ for ideal system

A

NEQ = SRNout^2=SNRin^2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

what angle is needed for FBCT

A

180 degrees plus fan angle

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

what is sinogram

A

radon transform of object
2d display of all projection profiles

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

Projection slice theorem

A

1D FT of a projection at angle φ(one row in sinogram)
= 2D FT of image evaluated along a radial profile at angle φ

Direct FT reconstruction algorithm:
1. Take 1D FT of the 1st projection angle in sinogram
2. Interpolate and add to a 2D rectangular grid F(kx,kv)
3. Repeat for all other angles in sinogram
4. Take 2D inverse FT of F(kx,kv) to find the image f(x,y)
cos, sin(,)(,)xryrrxykkkkPkFkkφφφ===
Q1. Is it the same as backprojection method?
Q2. Will we get 1/r blurring artifact?

A1. Note that direct FT reconstruction is not a backprojection technique.
A2. Unlike backprojection, direct FT reconstruction does not have 1/r blurring artifact.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

blurring introduced by backprojection

A

1/r blurring

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

filtered backprojection

A

amplify high frequency component by ramp filter

  1. Take 1D FT of 1st projection angle in sinogram
  2. Multiply this by ramp filter
  3. Take the inverse 1D FT
  4. Backproject the modified (filtered) projection
  5. Repeat for all angles around the object

multiplication o fpourier is faster than convolution in space

ramp also amplifies noise… (ie better resolution but more noise)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

why arc shape detector in CT?

A

avoid IS law

42
Q

HU-why normalize to uwater?

A

reduces energy dependence of CT value

43
Q

why are bowtie filters teflon?

A

zeff=8.4

44
Q

Same scan but reconstruct with:
A.40 x 5 mm slice, or
B.320 x 0.625 mm slice
Q. why (&how much) noise ↑ in B?

A

Since 8 times (5/0.625) thinner slice in B → photons ↓ by 8X → noise ↑ by sqrt(8) = 2.8X

45
Q

full fan vs half fan rotations

A

full fan goes 180+ fan angle
half fan goes 360 because detectors is shofted to the side to have larger FOV

46
Q

MIP vs min IP vs Ave IP

A

Max intensity projection (MIP):max value of a pixel over all phases
⇒indicates any location where tumour is present at least in one phase
Min-IP:min value of each pixel over all phases
⇒indicates only locations where tumour is constantly present over all phases
Ave-IP:mean value of each pixel over all phases
⇒indicates the time weighted location of the tumour

47
Q

when are images acquired in crdiac imaging?

A

end diastole- motion is minimal

48
Q

how dos ring artifact look in sinogram image?

A

In sinogram the artifact looks like a vertical line corresponding a fix radial distance from the center for all viewing angles.

49
Q

single ring artifact vs multiple ring

A

Single ring artifact: single detector faulty
Multiple ring artifact: improper CT calibration

50
Q

is ring artifact always du to detector?

A

No, eg if there is a small debris sticking to tube, bow-tie filter, detector face, or any thing on the beam path, it would cause a ring artifact, since we have a faulty ray.

51
Q
A
52
Q

why are effective doses in body larger than head?

A

larger tissue weighting factors in the body compare to head.

53
Q

CTDIvol vs CTDIw

A

CTDIw = 2/3 CTDI periphery + 1/3 CTDI center
CTDIvol = CTDIw/pitch
DLP= CTDIw times scan length

54
Q

impacting of averaging on SNR

A

SNR improves by square root of number of averages

55
Q

stable nuclei

A

isotopes:Nuclides with the same atomic number (Z)
isotones:Nuclides with the same no of neutrons (N)
isobars:Nuclides with the same mass atomic number (A)
isomers:same A,Z,N, different nuclear energy states (exited)

56
Q

avogrado number

A

6.02*10^23

57
Q

T1/2 for meytastable or isomeric state

A

T1/2= 10-9s to 30 years

58
Q

half lives of decay products of Mo99

A
59
Q

specific activity of neutron-ativation produced nuclides

A

radionuclide is an isotope of target →no chemical separation →specific activity ↓

60
Q

is equilibrium reached if parrents T1/2 < daugther T1/2?

A

no
In this case, daughter will reach a max activity and then decreases. Eventually, after parent almost completely decayed, the daughter will decay with its own half-life.

61
Q

when to elute moly cow?

A

After equilibrium Tc-99m activity is less than activity of Mo-99. This is because 12% of the Mo-99 decays directly to 99Tc without going through isomeric decay first.

62
Q

summary of radionuclide production methods

A
63
Q

alpha particle travel in air

A

1 cm/MeV in air
<1 mm in tissue

64
Q

where is antineutrino emitted

A

with beta minus (same side of arrow, or else it is neutrino on other side)
neutrino is with beta plus (same side of arrow, or else anti-neutrino on other side)

65
Q

how many ion pairs does a 1 MeV beta create?

A

a single ion-pair requires ~34 eV
eg. a 1-MeV βparticle creates:
106/34=3x104ion-pair=3x10-15C

66
Q

detection efficiency of ion chambers

A

< 1 %
Detections efficiency is low because the interacting medium is gas (low density)

67
Q

semiconductor detectors in nuc med

A

rarely used because DQE is low; however they have better energy resolution

68
Q

NaI properties

A

-most common in Nuc Med
sealed to avoid moisture, extraneous light
30% of crystal light reaches cathode
XFragile
XHygroscopic
Xγ>250 keV (Compton)
√Relatively efficient scintillator:one light photon per 30 eV
√Relatively high ρ, Z⇒absorbs most γ[50-250 keV] (photoelectric)
√grown in ovens to 30x50cm, 1cm thick
√Transparent
√Good wavelength matching

NaI(Tl) has a very high conversion efficiency (13% of deposited energy is converted in light), and 30% of produced light reaches the photocathode.

69
Q

PET scintillators

A

511 keVfrom positron emitters ⇒need ρ↑, Z↑
BGO:ρ↑, Z↑, common in PET
LSO:cost↑, less efficient
but brighter, faster
LYSO:similar to LSO

70
Q

why is it that even when nuc med detector is irradiated with mono-energetic spectrum the FWHM is not zero

A

statistical variations in detection process that causes the amplitudes of the pulses randomly vary about the mean pulse height.

71
Q

what nuclide is used for imaging, I-131 or I-123

A

I-123
erduced patient dose

72
Q

lower limit of activity measured by well counter

A

Can measure activities < 1 nCi

73
Q

confidence interval ranges

A

99.7% within 3 standard deviations
100.95% within 2 standard deviations
101.68% within standard deviation

74
Q

ideal sensitivity of nuc med detector

A

unity, i.e. 60 cpm/Bq

75
Q

minimum detectable activity

A
76
Q

What factors affect the sensitivity of the detector?

A

(1) intrinsic efficiency (1-exp(mu.x))
(2) geometric efficiency
(3) photofraction (fraction of the counts within photopeak , if only photopeak events are counted)
(4) absorption and scatter within source itself or material between source and detector

77
Q

where is diverging collimator used?

A

to image large objects with smaller detector

78
Q

where is diverging collimator used?

A

to image large objects with smaller detector

79
Q

where is diverging collimator used?

A

to image large objects with smaller detector

80
Q

effect of magnification on gamma camera spatial resolution

A
81
Q

what is collimator efficiency in nuc med proportional to?

A

~ area of hole/length of hole
efficiency os parrallel hole colli is independent of object distance from collimator, but resolution degrades as distance increases area of holes/length of holes

82
Q

how often do we test uniformity for anger camera

A

Uniformity tests are done daily.
Intrinsic uniformity measurement (without collimator in place) is done using a point source (150 micro-Ci Tc-99m) placed far away from the camera (> 4-5 times larger than the dimension of the crystal).
System uniformity measurement (with collimator in place) is done using a planar source (5-10 mCi Co-57) in front of the camera.
Lead (Pb) slit mask phantom has Pb sheets with thin slits.

83
Q

causers of spatial non-uniformity in nuc med

A
  1. Spatial non-linearity
    1. Interaction location affects energy (Z) signal
    1. Local variation in efficiency of absorbing γrays
84
Q

system efficiency for anger camera

A
85
Q

coincidence timing window

A

coincidence timing window (6 -12 nsec):max time diff to call it “in coincidence”
timing resolution of detector (0.5 -5 nsec):precision of coincidence window
-Typical energy window to accept a count for PET is 350 - 650 keV

86
Q

Is TOF PET more useful for pediatric or adult?

A

Time of flight improves the reconstructed signal to noise ration of large patients but not much useful with small patients such as pediatric

87
Q

what does non-colinearity of PET photons add to resolution

A

0.0022* diameter of body where anniholation can occure (2mm for 80 cm)

88
Q

typical detector element size in pet

A

3-6 mm

89
Q

what effect do septa have on scatter and random events in PET?

A

The use of septa (ie. 2D PET) will reduce both scatter and random events.

90
Q

PET scatter to true ratio as activity injected increases

A

The ratio of scatter-to-true does not increase with more injected activity, because trues and scatters both increase.

91
Q

pros and cons of 3D vs 2D PET

A

3D acquisition:
Xactivity outside FOV ↑
Xscatter ↑
Xdead time count losses ↑
Xrandom coincidence ↑
√true coincidences ↑
√sensitivity x4 -x8 ↑

Due to improved SNR in 3D PET, a higher cut-off frequency can be used in the reconstruction filter. Doing this can lead to an improved spatial resolution.

92
Q

standard uptake value if uniform uptake?

A

1

units of SUV are mCi/ml /(mCi*kg)

93
Q

why use multiple anger head cameras?

A

multiple head cameras (dual-head or triple-head) are used in SPECT to allow simultaneously acquisition at different angles.
Therefore, a multiple head camera can acquire the same number of counts in 1/2 or 1/3 of the time (compared with a single-head camera)

94
Q

how to get u map in nuc med?

A

use CT or use external source (Gd) to get transmission scan

For simultaneous acquisition, two separate energy windows must be used one for emission scan, one for transmission scan.

95
Q

when do you correct for scatter in nuc med?

A

Scatter corrections must be done before attenuation correction, otherwise the scatter events are also amplified by attenuation correction factor

scatter correction increases noise due to propagation of errors

96
Q

mean life in nuc med

A

1/lambda

97
Q
A
98
Q
A
99
Q
A
100
Q

wr for all radiation used in nuc med

A

1

Imaging Review CCPM (11 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions