II.

‘:::

The AAPM/RSNA

Physics
for Residents

: : Tutorial
This

artide

meets

for

criteria

hour

Clinical
Principles1

the

Applications

of Basic

X-ray

Physics

1.0 credIt

in category

Beth

A. Scbueler,

PhD

The

1 of

application

of basic

the AMA
Physician’s
Recognition
Award.

To obtain

credit,


see

the questionnaire
pp

quires

on

consideration

any

725-730.

given

each

radiographic

of these

current,

LEARNING

After


reading

(focal

and

taking

reader
.

selection
how

of x-ray

factors
density

and

Be able

the

I

to identify

by


technique

equipment

An

is

principles

and

factors.

Be able
basic

application.
.

Be able

to compare

scatter reduction
mechanisms
with respect

various


to image

patient

technique

measures

(source-object
patient

dose

contrast,

density,

mo-

patient

expo-

involves

quality

and


parameter

minimizing

while

of image

factors

of exposure

including

often

radiation

consideration

of

and exposure.

quality

dose.

and


#{149}
INTRODUCTION

in a
how

to identify the
types of x-ray
generators
and to select
the appropriate
type for a
particular
radiographic
.

quality,

of

geometry

of

tube

equipment

influence


for evaluation

unsharpness,

of radiographic

and

For

application

characteristics

also

re-

voltage,

In addition,
design)

of image

and geometric
various

basic


distance)

The basis

and

factors-tube

receptor.

generator

optimization

between

exposure

receptor

of the radiograph.

Selection

trade-offs

x-ray

radiography
interrelationships.


understanding

the

and image

use,

complex

contrast.

of unsharpness
radiograph
and learn
the visibility of detail

four

grid

is the

The

to clinical

have


time-determine

source-image

unsharpness,

sure.

image

causes

affected

tion

exposure

influences

size,
and

the quality

will:

Understand

exposure


spot

principles
that
proper

is essential.

to the patient

distance

article

the test, the

selection

.

tbis

physics
factors

examination,

factors


and

exposure

OBJECTIVES

x-ray

of many

understanding

production

of basic

and

x-ray

interaction.

to produce

physics

includes

Radiographic


an

image

that

knowledge

imaging
allows

the

of the

involves

principles

application

radiologist

of x-ray

of these

to visualize

the


basic

internal

anatomy
of the patient
so that a diagnosis
can be made.
For each radiographic
examination,
the operator
has control
over several
parameters
that affect
the appearance
of
the image,
including
the x-ray tube voltage,
tube current,
exposure
time, and distance.
In addition,
the design
of the radiographic
equipment
(x-ray tube and generator)
and

properties
of the particular
patient
and examination
(tissue,
contrast
media
used,
and
motion)
affect
radiographic
image
quality.
An understanding
of how x rays are produced
and interact
in tissue
is essential
to determine
the appropriate
selection
of technique
factors
and equipment
design
for a particular
clinical
examination.
The appearance

of a radiograph
is described
by various
image
quality
elements,
which
include
image
density,
contrast,
blur, and noise.
These
factors
describe
various
characteristics
themselves
ally

Index

has

RadloGraphics
the

RSNA

©RSNA,


Physics
1998;

Physics

ceived

that provide
interrelated

a detrimental

terms:

‘From

are

Department

November

effect

on

the

other


and comparing
in one image

factors.

Image

images.
The
quality
factor

evaluation

must

also

55905.

From

factors
gener-

include

‘ Radiography


18:731-744
of Diagnostic

Tutorial

a means
for evaluating
so that an improvement

at the

24; accepted

1996

Radiology,
RSNA

November

scientific
28.

Mayo

Clinic,

assembly.
Address


reprint

200

First

Received
requests

St, SW,
March

Rochester,

MN

1 3, 1998;

revision

to the

requested

the

AAPM/

May


27 and

re-

author.

1998

731


Figure
1. Radiographic
strate
how radiographic
was acquired
at 70 kVp

density
and the 10-kVp
rule. (a-c)
Lateral
density
varies
with different
milliampere-second
and 16 mAs. With the kilovoltage
unchanged,

with half the milliampere-seconds

the milliampere-seconds
or 32 mAs.

obtained
twice

diographic
density
and 16 mAs. With

10-kVp

reduction

varies
with different
the miffiampere-seconds

to 60 kVp,

or 8 mAs,

radiograph

exposure
as tow as possible.
article
examines
compromises
associated with the choice

of various
technique
and
equipment
design
factors
to provide
a guide
for
proper
selection
of x-ray technique,
focal spot
size, x-ray generator
type, and scatter
rejection
method.
Several
standard
texts provide
addiThis

information
and

on

image

principles


quality

of basic

x-ray

(1-7).

a SELECTION

OF X-RAY EXPOSURE
FACTORS
X-ray exposure
factors
include
the peak tube
voltage,
tube current,
and time that are selected
on the control
panel
of the x-ray machine
to produce
the desired
radiograph.
The
selection
of these
factors

affects
the image
density

and

tient

contrast

of the

radiograph

and

the

pa-

exposure.

. Control
of Image
Density
The primary
control
of image
density
ing of the radiograph)

is the product
current

and

exposure

time,

expressed

(blackenof tube
as milli-

ampere-seconds
(mAs).
Increasing
the mifiiampere-seconds
wifi proportionally
increase
the
number
of x rays that reach
the patient
and the
image
receptor.
Changes
in the tube voltage
also


affect

density,

but

in this

the tube voltage
will greatly
sure to the patient
and the
rays

through

changes
image

the

patient.

case
increase

increasing
the


transmission
As a result,

in tube voltage
cause
density.
The relationship

and

by observing
that the
overexposed
radiographs

patient

physics

phantom
demonradiograph
(b)
radiograph
(a) was

normal

Lateral

consideration

of the radiation
exposure
to the
patient.
Unfortunately,
image
quality
improvements
frequently
result
in greater
patient
exposure. Therefore,
it is important
to consider
ways to optimize
image
quality
while
keeping

tional

of a skull

values.
The
the underexposed

the overexposed

radiograph
(c) was obtained
at
radiographs
of a skull phantom
demonstrate
how rakilovoltage
values.
The normal
radiograph
(e) was acquired
at 70 kVp
unchanged,
the underexposed
radiograph
(d) was acquired
with a

(d-f)

and the overexposed

kVp. The 10-kVp
rule is demonstrated
are similar and the densities
of the

radiographs

expo-


of x
small

large changes
in
between
mdli-

(0

was acquired

with

a 10-kVp

densities
of the underexposed
(c, f) are similar.

ampere-seconds

and

tube

increase

voltage


that

#{149}maging
I

& Therapeutic

Technology

(a,

d)

results

in production

of equivalent
image
density
is
known
as the 10-kVp
rule: An increase
of 10
kVp is equivalent
to doubling
the milliampereseconds.
Figure

1 shows
how kilovoltage
and milliampere-seconds
can be manipulated
to change
image density.
As predicted
by the 10-kVp
rule, a
decrease
of 10 kVp produces
an image
with
density
similar
to that achieved
by reducing
the
milliampere-seconds

from

16 mAs

to

8 mAs,

and an increase
of 10 kVp produces

with density
similar
to that achieved
bling the miuiampere-seconds
from

an image
by dou16 mAs to
32 mAs.
It should
be noted
that the 10-kVp
rule
does not apply
for radiographs
acquired
at <60
kVp or >100 kVp or of small body parts such as
the extremities.
. Tube
Voltage
Selection
Selection
of tube voltage
is the primary
of controlling
contrast
in a radiograph.
contrast


is defmed

as the

difference

method
Image
in radio-

graphic
density
of adjacent
anatomic
structures.
The formation
of image
contrast
depends
on two independent
factors:
film contrast
and
subject
contrast.
Film contrast
depends
on the
characteristics
of the film used and how it is

processed,
which
is described
by the characteristic
curve.
Subject
contrast
is defmed
as the
relative
radiation
intensities
of the x-ray beam
exiting

the

patient.

The

subject

contrast

is

larger
if x-ray penetration
through

an object
is
much
different
from the penetration
through
adjacent
background
tissue.
The penetrability,
or penetrating
power,
is determined
by the effective
energy
of the x-ray beam:
Higher-energy
x-ray beams
penetrate
matter
farther
than towenergy

beams

do.

Because

x-ray


beam energy

directly
affected
by changing
the tube
the latter
is a major
factor
in determining
graphic
contrast.

732

to 80

radiographs

Volume

is

voltage,
radio-

18

Number


3


.

a.

d.

b.

-

c.

May-June

f.

1998

Schueler

U

RadioGrapbics

#{149} 33
7



-

.5.

...

4

Figure
2. Effect
torn was acquired

of tube

voltage

quired
at 100 kVp
in a large reduction

and 9 mAs.
in patient

at 70 kVp

is 1 1 5 mR

370 mR


Figure

tem

3.
shows

anatomic

(0.297
x 10’ C/kg),
(0.955 x io- C/kg).

Characteristic
curve
change
in film contrast
areas

with

different

on contrast

and

60 mAs.


In addition
exposure.
whereas

with

relative

represented
by the solid and dashed
ference
in optical
densities
between

Low-contrast

to a reduction
The entrance
the

for a screen-film

and dose.

(b)

skin

exposure


(a)

High-contrast
radiograph
of a skull phanradiograph
of a skull phantom
was ac-

in contrast,
skin exposure

produced

the

increase

in kilovoltage

for the low-contrast
in the

high-contrast

radiograph

is

Shoulder


sys-

exposure.
attenuation

results

radiograph

Two
are

lines. A larger
the two areas

dii-

indicates higher
contrast
is present
in the image.
When
the
anatomic
areas are properly
exposed,
the optical
densities fall within
the linear

portion
of the characteristic
curve
and the contrast
is greatest.
If the anatomic
areas
are overexposed,
the optical
densities
fall within
the
shoulder
portion
of the curve
and contrast
is reduced.

C,)

0

C)

C.

0
Use

of high


tube

voltage

results

in a reduc-

tion in contrast,
compared
with that achieved
with low kilovoltage
techniques.
This effect
is
demonstrated
by the two radiographs
in Figure
2. The image
obtained
at 100 kVp has substantially reduced
contrast,
compared
with that
seen in the 70-kVp
image.
The loss of contrast
results


in a decrease

in the

visibility

of detail

log Relative

Exposure

in

areas such as the frontal
sinus.
However,
when
the milliampere-seconds
is adjusted
so that the
amount
of radiation
reaching
the image
receptor is the same,
the 100-kVp
technique
requires


nique.
more
beam

a much

. Milliampere-Seconds
Selection
Selection
of milhiampere-seconds
affects
image
density,
as demonstrated
in Figure
la-ic.
In addition,
milliampere-seconds
selection
influences
contrast
in a secondary
way. For screen-

compared

734

.


Imaging

lower
with

radiation
that

& Therapeutic

exposure
needed

in the

Technology

to the
70-kVp

patient
tech-

The higher
penetrating,
is absorbed

kilovoltage
x-ray beam
is

so a smaller
fraction
of the
by the patient.

Volume

18

Number

3


.

4’..J
.

1)‘

a.

b.
Figure
4.
Loss of contrast
due to improper
exposure.
Underexposed

radiograph
of a skull
phantom
acquired
at 70 kVp and 30 mAs (a)
and the overexposed
radiograph
acquired
at 70
kVp and 1 20 mAs (c) are lower in contrast
cornpared
with the normal
exposure
acquired
at 70

kVp and 60 mAs (t).

the

difference

anatomic

in optical

areas

wilt


be

densities
the

between

largest

est contrast.
Film contrast
posure
results
in densities

for

the

two
high-

is reduced
when
exthat lie in the toe or

shoulder
regions
of the curve.
The effect

of Under- and overexposure
on contrast
in a clinical
image
is demonstrated
in Figure
4.

. FOCAL
SPOT SELECTION
The choice
of focal spot size primarily
influences
the amount
of geometric
unsharpness
a radiograph.
However,
focal spot selection

C.

film radiography,
both
underexposure
(mifiiampere-seconds
too low) or overexposure
(millampere-seconds
too high)
result

in a reduction
in film contrast.
The relationship
between
film
contrast
and density
can be understood
in
terms
of the characteristic
curve
(Fig 3). The
characteristic
curve
(or Hurter
and Driffield
[H&D]
curve)
describes
the relationship
between
optical
density
and exposure.
The curve
has

three


regions

that

correspond

1998

influences

the

amount

of motion

blur

in an

image,
since
tube current

the selection
limits the maximum
and tube voltage
settings,
thereby


affecting

exposure

the

time.

In addition,

design
of the x-ray
tant consideration,

tube anode
is also
because
the anode

may

focal

influence

the

field coverage,
vided
by the


spot

and heat
x-ray tube.

size,

capacity

the

an imporangle

radiation

that

are

pro-

to different

exposure
levels.
For low- and high-exposure
levels,
the slope
of the curve

is relatively
small.
These
portions
of the curve
are the toe and
shoulder
regions.
In between
the toe and
shoulder,
the curve
is a straight
line with
a
steep
slope.
Within
the straight-line
portion,

May-June

also

in

.

Focal


Spot

The x-ray focal
rays.

Instead,

size.

This

Blur
spot is not

a point

it is a rectangular
causes

a point

Schueler

source
region

in an

object


of x
of finite

to appear

#{149} adioGrapbics
R

U

735


F
Focal

Spot

I

SOD

SID

6a.
Object

Plane


OlD

1

Image Plane

Bf
5.

Figures

5, 6.
how

(5) Focal spot blur. Diagram
the focal spot blur in the image
_____________
plane (Bj7 increases
as the object is moved
6b
closer to the focal spot. OlD = object-image
distance,
SID
= source-image
distance,
SOD = source-object
distance.
illustrates

(6) Effect of focal spot size and magnification

on blur. (a, b) Radiograph
of the sella turcica,
obtained
with a small focal spot of nominal
size 0.3 mm (measured
size, 0.5 mm) (a), exhibits
greater
detail than
the radiograph
obtained
with a large focal spot of nominal
size 1.0
mm (measured
size, 1 .8 mm) (b). Both a and b have the same magnification
(M = 2). (c) Radiograph
obtained
with a large focal spot of
nominal
size 1 .0 mm (measured
size, 1 .8 mm) but with the object
in
contact
with the image receptor
(M = 1 . 1) is relatively
sharp cornpared
with b, even though
a large focal spot was used.

ing Bf
blurred


the
the

on

the

image.

The

amount

of blur

by

where

SID

the

magnification

M (M

=


source-image

distance):

Bf0
=

=

Bf/M

=Fx(1

R

Imaging

& Therapeutic

SID/SOD,

F x (OlD/SOD),

where
F = focal spot size, OlD = object-image
distance,
and SOD = source-object
distance.
To
compare

the focal
spot
blur to the size of the
object
itself, we calculate
the blur in the plane
of the object
(Bf).
Bf0 is determined
by divid-

736

=

in

image plane (BJ7 can be calculated
from
two similar
triangles
shown
in Figure
5:
Bf

_____

Technology


- 1/il).

The effect
of focal spot size and magnification on blur in a clinical
image
is demonstrated
in Figure
6. For the same magnification,
the focal spot
blur wifi increase
as the focal spot size
increases
(Fig 6a, 6b). In Figure
6b, bone
margins are indistinct
and some fme structures
blend into the background.
In addition
to the

Volume

18

Number

3


0.7


0.6

Composite

0.6

-

______

0.5

--..---.-

1.0 mm Focal
0.5

Spot

0.4
E
E

0.4

0.3

I-


0.3
0.2
0.2

Detail

Screen

0.1
0
1

1.2

1.4

1.6

1.8

2

2.2

2.4

1

1.2


1.4

Magnification

2

2.2

2.4

Magnification

Figure
7. Blur in the object
plane
as a function
of magnification.
cal spot and a high speed
screen
has a minimum
composite
blur
1 .0-mm
focal spot and a detail screen
has a minimum
composite

spot

blur also depends

on the
6c). When
there
is no magnification
(M = 1), the focal
spot blur is zero. if
magnification
is increased
by either
moving
the
object
away from the receptor
or moving
the
focal spot closer
to the object,
the focal spot
blur will increase.
Blur due to the image
receptor
will also
contribute
to the total image
blur in a radiograph.
Receptor
blur is primarily
caused
by the
spreading

of light photons
formed
by x rays interacting
with the intensifying
screen.
Because
the spreading
of emitted
light increases
as the
distance
between
the x-ray interaction
and film
increases,
the amount
of blur depends
on the
magnification

thickness

size,

the

(Fig

of the


screen

phosphor

layer.

A

thick,
high-speed
screen
has an inherent
blur
(Br) of approximately
0.7 mm,
whereas
the
blur from a thin,
detail
screen
is 0.2-0.3
mm.
As with focal spot blur, it is more
clinically
retevant
to calculate
the amount
of blur in the ohject plane
because
it can be compared

with the
size of the object
itself. The receptor
blur in
the object
plane
(Br0)
is determined
by dividing the inherent
blur in the image
plane
by the
magnification:
Br

=

Br/M.

1998

(a) A radiographic
system with a 1 .0-mm
at a magnification
of 1.5. (b) A system
with
blur when
no magnification
is used.


foa

the sum of the two components
squared.
The
contribution
to B from the two sources
depends
on the magnification
of the object.
As
magnification
increases,
focal spot blur increases
while
receptor
blur decreases.
The relationship

between

total

image

fication
can be demonstrated
cal spot-receptor
combination
graph

(8). Figure
7a shows

blur

and

magni-

for a particular
foin the form of a
the composite
im-

age blur for a radiographic
system
when
a 1.0mm focal spot and a high-speed
screen
are
used.
The curve
representing
focal spot blur
shows
how geometric
unsharpness
increases
with magnification.
The curve

representing
receptor blur shows improvement
in detail
with
magnification.
The composite
of the two mdicates
that the total image
blur decreases
then
increases

with

nification

is used,

magnification.
receptor

When
the magnification
size becomes
the major

When
blur

little


mag-

dominates.

is large,
the
determining

focal
factor

spot
in

the total image
blur. For this system,
a magnification
of 1 .5 will produce
the sharpest
radiograph.
Figure
7b demonstrates
the blur-magnification
relationship
when
a detail
screen
and
1 .0-mm focal spot are used. The detail screen

results
in tess receptor
blur compared
with
that produced
by the high-speed
screen.
For
this

equation
shows
that receptor
blur in the
object
plane
wifi decrease
as an object
is magnified.
The total image
blur in a radiograph
(H) is a
composite
of the focal spot blur and the receptor blur. It is calculated
as the square
root of
This

May-June


1.8

b.

a.

focal

1.6

system,

produced
ing the
receptor
tance.

the

sharpest

if magnification
object
as close
and increasing

Schueler

radiograph


wifi

be

is minimized
by placas possible
to the image
the source-image
ills-

#{149} adioGrapbks
R

#{149} 37
7


. Motion
Blur
Another
component
tal image

blur

that

minimized
by using
possible.

However,
factors
to produce
may

result

contributes

is patient

in an

to the

motion.
shortest

the

Motion
exposure

the selection
the shortest

increase

to-


blur

is

Focal

of technical
exposure
time

in focal

spot

Spot

blur.

Rotating

Anode

We

have seen that geometric
unsharpness
is decreased
by using a small focal spot. A small focal spot concentrates
heat on a smaller
area of

the anode
and results
in a tower
heat capacity.
A tow heat capacity
limits technique
settings
to
low power
or tow tube voltage
and tube current. With technique
selection
limited
to low
values,
exposure
time must to increased
to produce
adequate
image
density.
A large focal spot
can be used with higher
tube voltage
and tube
current
settings
for a shorter
exposure
and minimized

motion
blur.
In clinical
practice,
the compromise
tween

geometric

unsharpness

and

Track

time

Anode
Angle

I
I
I
I

time

Cathode

bemotion


blur

can be handled
by tailoring
the focal spot
lection
to the requirements
of the particular
amination.
When
image
detail
is important,

se-

small

focal

focal

spot

spot

blur.

When


large

focal

spot

should

tube voltage
the exposure

be

motion
with

values
time.

blur

higher

should

used

to


reduce

is a problem,
tube

be used

current

exthe

I
I
I’.

the

Focal
Size

I

and

Figure

to minimize

ing


8.

. Anode
Angle
The surface
of an x-ray tube anode
is angled
with respect
to the central
axis of the x-ray
beam
(Fig 8). Tubes
are produced
with anode
angles
that range
from 7#{176} 20#{176}.
to
This
angulation permits
larger
heat loading
while
minimizing the effective
focal spot size, which
is the
size of the x-ray source
as viewed
from the image. The angled
surface

increases
the width
of
the anode
focal spot track,
which
is defined
by
the surface
area impacted
by electrons
from
the filament
as the anode
rotates.
As a result,
the amount
of anode
angulation
influences
the
heat capacity
of the x-ray tube.
In addition,
the
anode
angle determines
the size of the area

Diagram

assembly

anode

tween

covered
by the x-ray beam,
since
the edge
the anode
wilt limit the angle of the emitted

Effective
Spot

the anode

depicts
the side view of a rotatof an x-ray tube. The angle be-

surface

and the central

field coverage
is limited.
An x-ray
duced
with an anode

with a large
a larger

area,

low
spot

track.

but

because

anode

rate

on

examination.

high
such

For

heat capacity
and
as cine angiography,


small

anode

angles

are

phy

requires

large

field

angles

are

of the
the

the

procovers

dissipation


width

practice,

depends

beam
angle

of heat

small

In clinical

angle

graphic

the

of the

ode

of

axis is de-

fined as the anode

angle.
The effective
focal spot
size is the length
and width
of the x-ray beam
projected
down
the central
axis.

is

focal

choice

particular

of

radio-

applications

requiring

small field coverage,
x-ray tubes
with

used.

General

coverage,

radiograso large

x

rays.

For a given
effective
focal spot size, the
choice
of the anode
angulation
is a compromise between
heat capacity
and field coverage.
As shown
in Figure
9, anodes
with small angles
provide
the highest
heat capacity,
but radiation


U GENERATOR
The x-ray generator
ity

and

technique
generator
promise

patient

SELECTION

design

exposure.

and focal
selection
of various

affects
Just

image

as with

#{149}maging

I

& Therapeutic

Technology

qual-

x-ray

spot selection,
the goat of
is to choose
the best comfactors

depending

on

particular
radiographic
examination.
The
basic types
of generators
are single
phase,
three
phase,
high frequency,

and constant

738

an-

needed.

Volume

18

the

four
po-

Number

3


Large
Anode
Angle

Small
Anode
Angle


I

Focal
Track

Spot
Width

Focal Spot
Track Width

/
/

/
/

I

/

I

I

I

/

I


I
Field

7

Coverage

Field

a.

Coverage

b.

Figure
9.
Comparison
size, the choice
of the

age. (a) Diagram
vides

a large

of small and large anode
angles.
For

anode
angulation
is a trade-off
between

depicts

focal

spot

the side view
track

width

of an anode

for a high

heat

with

a given
effective
heat capacity

a small


capacity,

angle.

but

the

focal spot
and field cover-

The small
resultant

angle

radiation

profield

coverage
is limited. (b) Diagram
depicts
the side view of an anode with a large angle. The
large angle provides
larger radiation
field coverage,
but the rate of heat dissipation
is low
because

of the small width of the focal spot track.

I

fsJ#{149}SstJ#{149}SsJ#{149}CsJ\

I

.

- -

-

.

Single

full wave

phase

rectified
(two pulse)

Three phase
Three pulse)
(six phase

I


. Patient
Exposure
and
Exposure
Time
Considerations
of patient
dose and exposure
time can be evaluated
by examining
the generator
voltage
waveform.
Figure
10 shows
representative
voltage
waveforms
for each generator type. The voltage
ripple
is defmed
as the
percentage

(twelve

pulse)

and


difference

minimum

single-phase

between

voltages

the

in the

generator

maximum

waveform.

exhibits

The

a 100%

voltage

ripple,

since the voltage
varies
from zero to the
peak value.
The three-phase,
six-pulse
generaHigh frequency

tor

has

a lower

Three-phase,

voltage

ripple

12-pulse

of

generators

1 3%-25%.
have

a ripple


Figure
10.
Diagram illustrates
representative
voltage waveforms
for single-phase,
three-phase
(sixpulse
and 12-pulse),
high-frequency,
and constant

of 3%- 10%, which
is similar
to the voltage
ripple
in high-frequency
generators
(4%-15%).
Constant
potential
generators
have no ripple.
A generator
with a large voltage
ripple
requires
a higher
patient

exposure
to produce
a
radiograph
at a certain
kilovoltage
selection.
These
types
of generators
produce
many low-

potential

energy

0,

-

Constant potential

0

>

Time

x-ray


generators.

tential.
Selection
criteria
of patient
exposure
and
exposure
reproducibility,
low unit cost.

May-June

1998

include
exposure
compact

minimization
time, good
size, and

x rays

that

do


not

contribute

age because
they are absorbed
Therefore,
the highest
patient
needed
when
a single-phase

Schueler

to the

by the
exposure
generator,

U

im-

patient.
is
which


RadioGraphics

U

739


has

100%

stant

voltage

potential

substantial

reduction

Generator
also

is used.

longer

in greater


Use

provides

in patient
with
large

types

require

sults

ripple,
generator

Primary
ripple

times,

blur.

most

dose.
voltage

exposure


motion

of a con-

the

This

X-rays

which

re-

is because

the

low-voltage
portion
of the exposure
pulses
does not deliver
a significant
exposure
to the
image
receptor,
thus the exposure

must be
lengthened
to produce
proper
image
density.
The total exposure
time required
when
a
single-phase
generator
used is longest
of the
four generator
types.
Applications
in which
rapidly
moving
structures
are imaged,
such as
cine angiography,
need
a generator
capable
of
producing
very short

exposure
times.
A constant potential
generator
is capable
of the
shortest

exposure

pulses

of approximately

J

111111111

Image

Receptor

msec.
An additional
generator
property
affecting
exposure
time is the generator
power

rating.
Use of a generator
with a higher
power
rating
allows
for selection
of higher
tube voltage
and
tube current
exposure
factors
so that exposure
times
can be shorter
for a desired
milliampereseconds.
. Exposure
Reproducibility
Good
exposure
reproducibility
produce
images
with uniform
reduce
the number
of retakes.
is also

phy

critical
because

for

digital

High-frequency

and

erators
provide
superior
to that
phase
generators.
three-phase

can
potential

reproducibility
with singleis because

output

the


ability

to compensate

for

variations

from

the

desired

that

discussion,

provide

the

conlowest

high

cost

of the

are

less

system.

High-fre-

much
more
compact
expensive.
In addition,

generators

can

be

in

designed

from either
singleor three-phase
line
age or from a battery
or charged
capacitor

for mobile
radiographic
units.

to

voltbank

and

directly

for

radiation

run

is

on
are

in-

time

sudden

line


voltage
changes.
High-frequency
generators
use closed-loop
regulation
to sense
the tube
voltage
and tube current
continuously
and to
correct

and

quency
generators
size and relatively

in-

that
or three-

the input
line voltage.
Voltage
regulators

cluded
in these
circuits,
but the response
limits

size

gen-

single-

depends

preceding

generators

high-frequency
constant

This

potential

of scattered

patient
exposure,
shortest

exposure
time, and
good
reproducibility.
However,
disadvantages
of constant
potential
generators
are the large

be-

cause

the amount
receptor.

. Size
and
Cost
As evident
from the

angiogravoltage

images

exposure
available


power

helps
reduce
reaches
the

is required
to
contrast
and to
Reproducibility

in tube

tween
mask and contrast
complete
subtraction.

11.
Cross-sectioned
diagram
shows
how a
grid placed between
the patient
and image receptor
Figure


stant

subtraction

differences

11111

0.5

settings.

U SCATI’ER
A large

fraction

undergo

REJECTION
of the x rays entering

Compton

interactions,

which

a patient

produce

scattered
x rays. The scattered
photons
are emitted in all directions,
but they tend to be scattered
in a more forward
direction,
as the energy
of the primary
beam is increased.
When
the primary x-ray beam
enters
hone surrounded
by soft
tissue,

the

radiographic

density

change

between

the soft tissue and bone should

be very large.
However,
the high contrast
is reduced
by scattered x rays, which
strike the image
receptor

740

U

Imaging

& Therapeutic

Technology

Volume

18

Number

3


a.

b.


Comparison
of grid and “nongrid”
techniques.
Both radiographs
of a skull phantom
were
with 90 kVp, 105-cm source-image
distance,
and 80-cm source-object
distance.
Radiograph
obtained
with a grid (grid ratio of 12:1 [grid thickness:interspace
width])
(a) demonstrates
a
noticeable
improvement
in contrast
compared
with the nongrid
radiograph
(b). In addition,
a substantial
increase
in patient
dose was required
for the grid radiograph
(150 mR [0.387

x 10
C/kg])
compared
with the nongrid
radiograph
(33 mR [0.085 x 10’ C/kg]).
Figure

12.
obtained

can

the shadow
of the bone.
Several
methods
be used to reduce
the amount
of scattered
x

The increase
in contrast
is achieved,
ever,
at the expense
of increased
patient


howdose.

rays

that

The lead strips
of the grid absorb
some
radiation
that would
have reached
the
receptor;
thus, an increase
in exposure
quired
to achieve
the same film density.
entrance
skin exposure
for the radiograph
taken
with the grid is 150 mR (0.387
x
kg), whereas
the skin exposure
produced
obtaining
the radiograph

without
a grid

of the
image
is reThe

stantially

C/kg).

within

reach

the

receptor.

oils are the use of grids
tion to limit the volume
reduces
the production

Two

of these

meth-


or an air gap. Collimaof irradiated
tissue also
of scattered
x rays.

. Grids
The most common
method
of reducing
the
level of scattered
radiation
reaching
the image
receptor
is use of grids (9). A grid is constructed
of alternating
strips oftead
and nonabsorbing
interspace
material
and is placed
between
the
patient
and image receptor.
The strips are arranged

a line
rected


to transmit

from
at an

the

only

x-ray

angle

are

those

source

x rays

(Fig

preferentially

directed

1 1). X rays
absorbed


in

1998

at 33 mR

(0.085

x 10

Cl
in
is sub-

The ratio of the exposure
required
with grid
use and without
grid use is called
the Bucky
factor.
The Bucky
factor
is higher
for higher
tio grids and higher
energy
exposures.


ra-

diby

the grid. Because
most scattered
x rays are emitted at an angle to the primary
beam direction,
a
large fraction
of the scattered
radiation
is absorbed.
Figure
12 demonstrates
the contrast
improvement
that can be obtained
by using a grid.

May-June

tower

10

. Air Gap Technique
Another
method
of reducing

tered
radiation
that reaches
is to place

a gap

between

Schueler

the level
the image
the

patient

U

of scatreceptor
and

RadioGraphics

the

#{149}
741



receptor
(10). Because
are emitted
at an angle

most scattered
x rays
to the direction
of the
primary
beam,
a large
fraction
will not strike
the receptor
if it is separated
from the patient
by

a sufficient

distance

(Fig

13).

However,

Primary

X-rays

pri-

mary x rays
source
wifi

directed
in a line from
the x-ray
not be affected.
The typical
air gap
distance
is 1 5-45
cm, which
wifi also introduce
some
magnification
and limit the field of
view of the subject.
The change
in contrast
that results
from use of an air gap is shown
in
Figure
14.


Patient

Clinical
Applications
Both grid and air gap techniques
are effective
means
of controlling
scattered
radiation
and
improving
contrast
in a radiograph.
To select

S

which
method
is best for a specific
application, we must examine
the trade-offs
involved.
Grids do not require
use of magnification,
so
focal spot blur is reduced.
When
an air gap is

used,
a small focal spot is generally
needed
to
minimize the geometric
unsharpness.
The advantages
of the air gap technique
include
use
of lower
milliampere-second
values,
compared
with that needed
for the grid technique,
which
results

in less

tube

tient exposure
may
gap, but the amount
the source-to-patient

loading.


In addition,

the

be reduced
by using
of reduction
depends
distance
used.

Air
Gap

$
Figure
13.
Cross-sectioned
an air gap placed
between

diagram
shows
how
the patient
and image
receptor
helps reduce
the amount
of scattered

radiation that reaches
the receptor.

pa-

an air
on

For most radiographic
examinations,
grid
use is common.
However,
there
are several
applications
in which
the air gap technique
offers
some
advantages
over grid use. One such application
is cerebral
angiography
(1 1). When
the
air gap technique
is used in cerebral
angiography, the geometric
magnification

is generally
adjusted
to 1 .5- 1 .8. Some
radiologists
prefer
the magnified
images,
which
are free of grid
lines. Compared
with a grid technique,
the increase
in patient
exposure
resulting
from positioning

the

patient

closer

to the

x-ray

source

approximately

the same as the exposure
increase
required
when
a grid is used. Another
air gap application
is chest
radiography
(12).
this

case,

because

a large

source-image

tance
(6 or 10 feet) is typically
used, the
an air gap can substantially
reduce
patient
and the resulting
magnification
is slight.

is


. SUBJECT
CONTRAST
The amount
of subject
contrast
produced
is affected by both physical characteristics
of the
object
ray

and

penetrating

beam.

ness,

Object

physical

number

(Z).

different


sorb

In

use of
dose

density,

Two

voltage
erator,

Imaging

& Therapeutic

Technology

atomic

have either
or Z will ab-

of radiation.

X-ray

tube


and

voltage

beam
the

waveform
produced
by the x-ray genplus filtration
(ie, added
and inherent
material

in the

path

of the

x-ray

beam).
As demonstrated
in Figure
2, kilovoltage has a substantial
effect
on image
contrast.

In addition,
factors
such as filtration
and voltage waveform
shape
also alter the distribution
of x-ray energies
in the beam.
The

way

in which

to produce

mined

#{149}

x-

thick-

that

densities,

include


gether

742

effective

areas

amounts

characteristics

of the
include

and

tissue

thicknesses,

different

attenuating

dis-

characteristics
characteristics


by Compton

these
subject

and

factors
contrast

photoelectric

Volume

come

to-

is deter-

interac-

18

Number

3


a.


b.

Figure
14.
Comparison
of air gap and “non-air
gap”
tom were
acquired
at 90 kVp without
a grid. Radiograph
strates
a noticeable
improvement
in contrast
compared
ample,
the source-image
distance
was adjusted
so that
proximately

the

Effective
Densities
Materials


same

to

better

compare

techniques.

Both

numbers,

and

ties

are

ton

very

Effective
Atomic
Number

Physical
Density

(g/cm3)

beam

1.00
1.00
0.91
3.50
4.93
0.0013

7.5
5.9

56.0
53.0

7.6

not
for

mize
ray

are

the

tissue


.
Compton

in tissue.

interactions

manly

on

dence

on atomic

tissue

occurrence
strong
ray

probability

occurring

density,

number


with

energy.

The

teractions

increases

as energy

on

of

x-ray

depen-

being

interactions
atomic

has

number

probability


and

a

of photoelectric

as Z increases
increases.

sity.

Radiography
tissue
requires

to radiographic

technique.

ids,

May-June

and

fat

1998


have

relatively

so that

The

special
Muscle,

attention

has

Because

tissue

effective

x-

of softin which

the

the

a


and

we

need

barium.

dennormally

material

maximize

Use

of 60-75

it produces

many

iodine

and

structures

contrast

to

is in-

contrast

number

of the

iodinated

the
object

contrast

used

atomic

size

into
of the

subject

iodine


absorption.
because

introduced

commonly

a high
the

with

above

most

contain

small,

able

low

imaged

imaged

de-


example

attenuation

usually

and

ential

. Soft-Tissue
Imaging
of soft

be

the

Iodine
in-

anode

can

materials

x-

maxi-


a tow-energy

Media

creased.

The

tube

to change

little

energy.

To

kVp are used with
and filter to produce

of 25-30

agents

body

very


An

inter-

types.

is mammography

Contrast

pri-

interac-

x rays.

Contrast

depends

or x-ray

of photoelectric
dependence

creases

The

tissue


used.

factors

when
x-ray

Compton

effect,

radiography

to occur

as photoelectric

distinguishing

technique

special

likely

However,

be


densiComp-

a high-energy

as effective

must

and

As a result,

more

photoelectric

beam

numbers

with

keY).

are

actions

7.4


of x rays

this exis ap-

(Table).

is imaged
(>40

tions

atomic

similar

low-energy

tions

(b). For
radiographs

their

interactions

a patient

Water
Muscle

Fat
Barium
Iodine
Air

of a skull phanair gap (a) demon-

contrast.

Atomic
Numbers
and Physical
of Human
Tissues
and Contrast

Substance

radiographs

obtained
with a 25-cm
with a contact
radiograph
magnification
in the two

kVp

K-absorption


are

the
x rays

edge

differ-

is preferjust

of 33 keV.

fluatomic

Schueler

U

RadioGraphics

I

743


x-ray

K-absorption

absorption

ergy

is equal

(The

edge is an abrupt
increase
that occurs
when
the x-ray

to or

energy

binding
The

atomic

slightly

greater

of the K-shell
number


than

in
en-

case,

normally

fluid-filled

cavities

are

233-266.
Bushberg
JT, Seibert JA, Leidholdt
EM, Boone
JM. The essential
physics
of medical
imaging.
Baltimore,
Md: Williams
Bushong
SC. Radiologic
gists: physics,
biology,


a
4.

5.

absorption

difference

from

the

application

clinical

large

6.

of x-ray

radiography

many

interrelated

technique,


scatter
factors

focal

rejection
generally

for

these

quality

without

to

7.

of

including

type,

8.

exposure


generator

size,

method.
involves

selecting

of image
posure.

principles
consideration

factors,
spot

should
be based
on the
tics of a given
radiographic
basis

physics

requires


and

Selection
of these
compromises
that

patient

9.

cx-

10.

P. Physical

principles

of medical

imag-

Va: American

College

of Radiology,

Bucky


1 1.

1 . Arnmann

E. X-ray generator
and AEC design.
In: Seibert
JA, Barnes
GT, Gould
RG, eds.
Specification,
acceptance
testing,
and quality
control
of diagnostic
x-ray imaging
equipment.
12.

G.

A grating

diaphragm

to cut

off sec-


ondary
rays from the object.
Arch Roentgen
Ray 1913; 18:6-9.
Gould RG, Hale J. Control
of scattered
radiation

REFERENCES

U

Sprawls

x-ray imPhysics,

1984.

is optimization

excessive

considerations

ing. 2nd ed. Gaithersburg,
Md: Aspen,
1993.
Sprawls
P, Lamel DA. Principles

of imaging,
Sequence
142 (case no. 8042). Diagnostic
Radiological Health Sciences
Learning
Laboratory.
Reston,

particular
characterisexamination.
The
factors

imaging

designs.
In:
film mamand mcdiMadison,
Wis:

physics
responsibilities.
Medical
Physics,
1991; 47-66.
Hasegawa
BH. The physics
of medical
aging.
2nd

ed. Madison,
Wis: Medical
1991.
cat

CONCLUSIONS

The

St Louis, Mo: Mosby,
1988.
Curry
TS, Dowdey
JE, Murry RC. Christensen’s
physics
of diagnostic
radiology.
4th ed. Philadetphia,
Pa: Lea & Febiger,
1990.
Gauntt
DM. Mammography
x-ray generators:

mography:

in density.

.


results

& Wilkins,
1994.
science
for technoloand protection.
4th ed.

conventional
and high frequency
Barnes GT, Frey GD, eds. Screen

filled

with air. Even though
the effective
atomic
number
of air is similar
to that of soft tissue,
differential

20. WoodPhysics,
1994;

of

3.

electrons.)

density
of barium

and

No.

2.

the

are similar
to those
of iodine,
but the size of
the structures
normally
imaged
with barium
contrast
agents
are generally
large. Therefore,
high kilovoltage
technique
is used
to penetrate
the contrast
agent
and better

visualize
the lumen.
Air can also be used as a contrast
agent.
In
this

Medical
Physics
Monograph
bury,
NY: American
Institute

by air gap

techniques:

application

to chest

radiography.
AJR 1974;
122:109-114.
Barnes GT, Feretti JM, Lamet DA. Principles
of
imaging,
sequence
1 27 (case no. 8027). Diagnostic Radiological

Health
Sciences
Learning
Laboratory.
Reston,
Va: American
College
of
Radiology,
1984.
Barnes

GT,

Fraser

RG.

Principles

of imaging,

sequence
128 (case no. 8028).
Diagnostic
Ridiological
Health Sciences
Learning
Laboratory. Reston,
Va: American

College
of Radiology, 1984.

This
Award.

744

U

hnaging

article

meets

To obtain

& Therapeutic

the
credit,

criteriafor
see

1.0
the

credIt


questionnaire

Technology

hour

in category
on pp

1 oftbe

AMA

Physician’s

Recognition

725-730.

Volume

18

Number

3