Medical Imaging

Published on February 2017 | Categories: Documents | Downloads: 35 | Comments: 0 | Views: 803
of 25
Download PDF   Embed   Report

Comments

Content


J
.

T
.

D
o
b
b
i
n
s

I
I
I
,

e
t
.

a
l
.
.

"
M
e
d
i
c
a
l

I
m
a
g
i
n
g
.
"
C
o
p
y
r
i
g
h
t

2
0
0
0

C
R
C

P
r
e
s
s

L
L
C
.

<
h
t
t
p
:
/
/
w
w
w
.
e
n
g
n
e
t
b
a
s
e
.
c
o
m
>
.
© 1999 by CRC Press LLC
íodIcuI ImugIng
79.1 Introouction
79.2 Imagc Inlormation Contcnt
Mcasurcmcnt ol Imaging Pcrlormancc
79.3 XPay Imaging
XPay Imaging Dctcctors
79.1 Computco Tomography
Pcconstruction ol an Objcct lrom Projcctions · Clinical
Mcasurcmcnts
79.5 Nuclcar Mcoicinc
Mcasurcmcnt ol Physiological Iunction · Mcasurcmcnt ol
Tcchnical Pcrlormancc
79.o Positron Fmission Tomography
Principlc ol Coinciocncc Dctcction · Dctcctor Compostion ·
PFT Scanncrs
79.7 Ultrasouno Imaging
Characcristics ol Souno Wavcs in Tissuc · ßMooc Imagcrs ·
Dopplcr Tcchniqucs · Color Dopplcr Imaging · Mcasurcmcnt
ol Ultrasouno Instrumcnt Pcrlormancc
79.8 Magnctic Pcsonancc Imaging
MP Imaging Tcchniqucs · MP Imagc Contrast · MP
Instrumcntation
79.1 IntroductIon
Mcoical imaging has aovancco consiocrably sincc thc oiscovcry ol Xrays by Wilhclm Conrao Pontgcn
in 1895. Tooay, in aooition to thc continuco usc ol Xrays lor mcoical oiagnosis, thcrc arc imaging
mcthoos that usc souno (ultrasouno), magnctic lclos ano raoio wavcs (magnctic rcsonancc imaging),
ano raoionucliocs (nuclcar mcoicinc). ßoth projcction imaging ano crossscctional imaging arc routincly
usco clinically. This chaptcr will ocscribc thc principlcs bchino thc various imaging mooalitics currcntly
in usc, ano thc various mcasurcmcnts routincly maoc with thcm.
79.2 Imagc InformatIon Contcnt
Thc vast majority ol imaging proccourcs arc qualitativc in naturc, whcrc it is thc visual prcscntation ol
anatomy that is thc mcasurcmcnt outcomc. Thcrc arc also somc quantitativc mcasurcmcnts, which will
bc oiscussco in thc scction on nuclcar mcoicinc. Howcvcr, sincc most imaging is conccrnco only with
thc qualitativc naturc ol thc imagc, a ocscription ol thc salicnt lcaturcs ol imagc contcnt lollows.
Thcrc arc thrcc primary physical paramctcrs ol intcrcst in imagc contcnt. contrast, noisc, ano rcsolu
tion. Il thcsc thrcc lcaturcs arc known lor a givcn imagc (or imaging systcm), thcn thc cntirc physical
naturc ol thc imagc has bccn charactcrizco. Thcrc arc also psychovisual cllccts, such as conspicuity {1],
]umos J. LohhIns III
IuIc !n!\crs!¡y Acd!ca! Ccn¡cr
Soun í. Humos
IuIc !n!\crs!¡y Acd!ca! Ccn¡cr
ßruco H. Husoguvu
!n!\crs!¡y oI Ca!!Iorn!a, 5an
Iranc!sco
JImoIhy R. LoCrudo
IuIc !n!\crs!¡y Acd!ca! Ccn¡cr
]umos À. ZugzohskI
!n!\crs!¡y oI V!scons!n, Aad!son
RIchurd Iruyno
!n!\crs!¡y oI V!scons!n, Aad!son
© 1999 by CRC Press LLC
which allcct thc ability ol thc obscrvcr to octcct a particular lcaturc, but thcsc issucs arc oillcult to
quantitatc ano arc outsioc thc scopc ol this hanobook.
Thc lrst ol thc physical imagc lcaturcs, contrast, is oclnco as thc lraction ol thc total imagc signal
occupico by a particular objcct.
(79.1)
whcrc S is thc signal in thc arca ol intcrcst ano ß is thc backgrouno signal. Contrast is octcrminco by
thc propcrtics ol thc objcct bcing imagco, thc imaging mooality, thc propcrtics ol thc imagc octcctor,
postproccssing ol thc imagc (such as by oigital proccssing), ano thc contrast ol thc oisplay ocvicc.
Imagc noisc is a mcasurc ol thc stochastic naturc ol thc imagc. All physical mcasurcmcnts, incluoing
mcoical imagcs, contain a ccrtain ocgrcc ol unccrtainty. In Xray imaging, lor cxamplc, thc physics ol
Xray proouction oictatcs that thc numbcr ol Xrays inciocnt on a unit arca pcr unit timc arc ranoom,
ano givcn by a statistical oistribution known as thc Poisson oistribution. Thc grcatcr thc imagc noisc,
thc lcss likcly it is that onc will obscrvc a givcn objcct. Thcrc is a rclationship bctwccn thc imagc noisc,
thc contrast ano arca ol an objcct, ano its likclihooo ol bcing obscrvco. This is summarizco in thc Posc
moocl.
(79.2)
whcrc N is thc numbcr ol quanta (such as Xrays) pcr unit arca nccoco to oisccrn an objcct ol contrast
C ano arca A, assuming a signaltonoisc ol |. Posc louno that a signaltonoisc ratio ol 5 is typically
rcquirco to octcct a visual objcct rcliably {2].
Pcsolution is thc ability ol an imaging systcm to rccoro laithlully thc rangc ol spatial octail in an
objcct. Pccoroing objccts with lncr spatial octail rcquircs "sharpcr` imaging octcctors. Thc rcsolving
ability ol a octcctor is largcly octcrminco by its pointsprcao lunction. Thc pointsprcao lunction
ocscribcs how wcll thc imaging apparatus can rccoro an inlnitcsimal point objcct. No octcctor is pcrlcctly
sharp, ano somc sprcao ol thc inlnitcsimal oot occurs thc worsc thc sprcao, thc lcss rcsolving thc
systcm.
Mcasurcmcnt of ImagIng Pcrformancc
A lincarsystcms approach is typically usco to quantily thc pcrlormancc ol an imaging systcm. Thc
rclations among contrast, noisc, ano rcsolution ol an imaging systcm arc customarily ocscribco by two
lunctions. thc mooulation translcr lunction (MTI) ano thc noisc powcr spcctrum (NPS), both ol which
arc lunctions ol spatial lrcqucncy. Thc MTI is thc Iouricr translorm ol thc pointsprcao lunction, ano
ocscribcs thc inhcrcnt octcrministic lrcqucncy rcsponsc ol thc systcm. Thc NPS (also rclcrrco to as thc
Wicncr spcctrum) is proportional to thc squarc ol thc Iouricr translorm amplituoc at cach lrcqucncy,
ano rcprcscnts thc variancc associatco with noisc in thc systcm at cach particular spatial lrcqucncy. Thc
ratio ol MTI ano NPS, propcrly normalizco, is thc noisc cquivalcnt quanta (NFQ), which is thc squarc
ol thc maximum availablc signaltonoisc at cach spatial lrcqucncy u.
(79.3)
Il thc NFQ is oivioco by thc numbcr ol inciocnt quanta pcr arca (c.g., thc numbcr ol Xray photons
inciocnt on thc octcctor in Xray imaging), thc rcsult is thc octcctivc quantum cllcicncy (DQF). Thc
C
ß S
ß

÷
N
|
C A

2
2
NIÇ u
u
u
,
¬
,

,
,
largc arca signal MTI
NPS
2
2
© 1999 by CRC Press LLC
zcrolrcqucncy DQF is a mcasurc ol thc lraction ol inciocnt quanta cllcctivcly usco by thc systcm.
Altcrnativcly, thc DQF may bc vicwco as thc cllcicncy with which thc systcm utilizcs thc availablc signal
tonoisc at cach spatial lrcqucncy.
Thc actual mcasurcmcnt ol MTI, NPS, ano DQF is quitc tcoious, ano will bc only bricly summarizco
hcrc. Thc intcrcstco rcaocr is cncouragco to consult thc suggcstco rclcrcnccs lor thc appropriatc octail
on thcsc mcasurcmcnts. Fxamplcs ol thcsc mcasurcmcnts will bc givcn lor Xray imaging.
Thc MTI is typically mcasurco by imaging cithcr a vcry lnc slit (typically 10 to 20 µm) {3o] or an
cogc {7]. Thc prollc across thc slit imagc is callco thc lincsprcao lunction (ISI). Thc Iouricr translorm
ol thc ISI givcs thc MTI in thc oircction pcrpcnoicular to thc slit. Thc ocrivativc ol valucs along thc
cogcrcsponsc lunction also givcs thc lincsprcao lunction. Dctcctor rcsponsc typically varics with cncrgy
so it is important to spccily thc conoitions unocr which MTI is mcasurco. With Xray imaging it is
typical to usc a tubc voltagc ol 70 kV with 0.5 mm Cu lltration placco in thc bcam to simulatc thc
lltcring ol thc Xray spcctrum cxpcctco lrom a paticnt, although othcr mcasurcmcnt tcchniqucs arc also
louno in thc litcraturc.
Thc NPS is mcasurco by taking an imagc ol a lat lclo, whcrc thcrc is no structurc in thc imagc othcr
than noisc. Contcmporary mcthoos ol NPS mcasurcmcnt on oigital systcms pcrlorm a twooimcnsional
Iouricr translorm on thc latlclo imagc {o], although whcn mcasuring thc NPS ol llm a scanning slit
is usco to gcncratc a oncoimcnsional NPS parallcl to thc oircction ol slit movcmcnt {810]. Altcr
appropriatc scaling, thc squarc ol thc amplituoc ol thc twooimcnsional Iouricr translorm is thc NPS.
Thcrc arc many octails rclatco to mcasuring thc NPS propcrly, incluoing climinating backgrouno trcnos,
ano thc sizc ol thc rcgion ovcr which thc Iouricr translorm is takcn. Thcsc arc all covcrco in octail in
thc rclcrcnccs {o,818].
Mcasurcmcnt ol imaging propcrtics is casicr on oigital imaging systcms than on llm, sincc llm must
lrst bc oigitizco at appropriatcly lnc sampling intcrvals or clsc corrcctco lor thc usc ol oncoimcnsional
slits {19]. Howcvcr, thc cllccts ol aliasing (lctitious lrcqucncy rcsponsc in a oigital systcm ouc to limitco
sampling) makcs thc intcrprctation ol MTI ano NPS in oigital systcms morc oillcult than with llm
{11,12,20].
79.3 X-Ray ImagIng
Xray imaging rcquircs an Xraygcncrating apparatus (tubc, high voltagc supply, ano controls) ano an
appropriatc Xray octcctor. Typical Xray octcctors incluoc photographic llm (almost always usco in
conccrt with a luorcsccnt scrccn), imagc intcnsilcrs, computco raoiography phosphor platcs, ano ncwcr
ocoicatco oigital octcctors.
Thc Xray gcncrator is basically a highvoltagc stcpup translormcr with appropriatc rcctilcation ano
control circuitry. Most contcmporary gcncrators arc thrccphasc 12pulsc, lullwavc rcctilco to givc a
vcry low voltagc ripplc (3 to 10%) {21]. Ior proccourcs rcquiring vcry last pulscs ol scvcral milliscconos
or lcss (such as coronary angiography), a tctroocbasco constantpotcntial gcncrator is usco. Thc opcrator
sclccts thc tubc kilovoltagc, tubc currcnt, ano cxposurc timc appropriatc lor thc cxamination ol intcrcst.
Xray tubcs contain a hcatco llamcnt (which scrvcs as thc cathooc) ano an anooc maoc ol a tung
stcn/rhcnium combination lor convcntional usc or molybocnum lor mammography. With thc cxccption
ol ocntal tubcs, moocrn clinical Xray tubcs almost always contain a rotating anooc to sprcao thc hcat
out ovcr a largcr arca, allowing lor a grcatcr tubc output without oamaging thc anooc. Many tubcs
contain two llamcnts, a largc onc ano a small onc, ocpcnoing on tubc output ano rcsolution rcquircmcnts
ol a particular cxam. Mcasurcmcnts on Xray tubcs ano gcncrators involvc calibrations to assurc that
kilovoltagc, tubc output, ano cxposurc timc arc in gooo agrccmcnt with thc control consolc scttings {21].
Calibration ol thc high voltagc is oonc by commcrcially availablc voltagc oiviocrs, or by spccially ocsignco
Xray llm casscttcs with calibrating lltcrs insioc. Tubc output is mcasurco by ion chambcrs, ano cxposurc
timc is mcasurco cithcr by a rotatingarm timcr tcst tool placco ovcr a llm casscttc ouring an Xray
cxposurc or by oircct plotting ol thc cxposurc vs. timc output ol an ion chambcr.
© 1999 by CRC Press LLC
X-Ray ImagIng Dctcctors
Thc most common octcctor lor Xray imaging is llm. Xray llm is typically placco in a sanowich bctwccn
two luorcsccnt scrccns (or onc scrccn in mammography lor improvco visibility ol small octail). Con
tcmporary scrccns arc maoc ol rarc carth compounos such as Co
2
O
2
S, ano scrvc to convcrt thc Xrays
to visiblc light which cxposcs thc llm morc cllcicntly than Xrays alonc, thus rcoucing paticnt raoiation
oosc. Thc rcsponsc ol thcsc scrccnllm combinations has gooo contrast at intcrmcoiatc cxposurc rangcs
(as givcn by thc llm y, or contrast ratio), but poor contrast at low or high cxposurcs. Thc contrast ano
latituoc ol llms arc ocscribco by thc charactcristic curvc (oltcn rclcrrco to as thc HurtcrDrillclo, or
HD curvc). Appropriatc scrccnllm combinations arc choscn basco on thc anatomy to bc imagco, sincc
scrccnllm combinations arc ocsignco with oillcrcnt contrast, latituoc, ano cxposurc scnsitivity charac
tcristics {22].
A sccono typc ol Xray octcctor is thc imagc intcnsilcr, which is usco with luoroscopy. Iluoroscopy
uscs a lowcxposurcratc Xray output to imagc a paticnt continuously, typically to propcrly position thc
paticnt lor a subscqucnt highcxposurc llm imagc. Thc imagc intcnsilcr (Iigurc 79.1) compriscs a
cylinorical glass cnclosurc, insioc ol which is an input scrccn, photocathooc, locusing clcctroocs, acccl
crating anooc, ano output scrccn {21]. Thc Xrays arc absorbco in thc input scrccn (typically CsI), giving
oll light which libcratcs clcctrons lrom thc photocathooc. Thc photoclcctrons arc thcn accclcratco to thc
output scrccn whcrc thcy strikc thc output phosphor scrccn with high cncrgy (~30 kcV), giving oll a
bright light, which is vicwco by cithcr a vioco camcra or motionpicturc (cincraoiographic) camcra.
A rcccntly ocvclopco oigital Xray octcctor is thc photostimulablc phosphor, which is rclcrrco to
commonly as computco raoiography {2325]. This octcctor uscs a spccial typc ol phosphor which storcs
about hall ol thc absorbco Xray cncrgy in mctastablc statcs, which arc rcao out latcr by lascr scanning.
Thc lascr light stimulatcs thc phosphor to cmit ultraviolct light in proportion to thc original Xray
cxposurc. Thc photostimulatco light is thcn octcctco with a photomultiplicr tubc (PMT) or soliostatc
photooctcctor ano oigitizco. Thc clinical apparatus (Iigurc 79.2) lrst oocs a prcscan ol thc imaging platc
to aojust thc input rangc ol thc analogtooigital convcrtcr basco on thc imagc histogram, thc oigitizco
signal is thcn logarithmically translormco ano storco, oisplayco on a vioco monitor, or printco on llm
lollowing optional spatial lltcring ano contrast aojustmcnt.
Thcrc arc also currcntly availablc or in ocvclopmcnt a varicty ol othcr oigital Xray octcctors, incluoing
sclcnium platc octcctors {2o], CCDcamcra octcctors with luorcsccnt scrccns {27], ano latpancl arrays
with amorphous silicon {28] or amorphous sclcnium {29] octcctor clcmcnts.
79.4 Computcd Tomography
A oiagnostic computco tomography (CT) scanncr compriscs an Xray tubc with collimation to provioc
thc slicc thickncss, a lincar array ol octcctor clcmcnts, ano a rcconstruction computcr. Thc Xray tubc
ano thc octcctors typically rotatc in a gantry. Thc numbcr ol octcctors usco ocpcnos on thc gcncration
ol thc scanncr. Iirstgcncration scanncrs hao only onc octcctor which was translatco across thc paticnt
with thc tubc lor cach projcction, ano thcn thc cntirc asscmbly was rotatco to acquirc thc ncxt projcction
vicw. To incrcasc acquisition spcco, scconogcncration scanncrs usco scvcral octcctors in a limitco lan
bcam gcomctry. Thirogcncration CT scanncrs, which arc thc most common in usc tooay, utilizc a largc
lan array ol octcctors (852 clcmcnts in a currcnt scanncr), which complctcly cncompasscs thc paticnt
ano allows slicc acquisition timcs ol about 1 s {21]. Thc Xray tubc ano thc octcctor lan array arc
mcchanically couplco ano rotatc togcthcr at high spcco (Iigurc 79.3).
Thc implcmcntation ol clcctronic slip rings, which allow continuous clcctric contact, has rcmovco thc
physical rcstriction imposco by thc highvoltagc cablcs ol carlicr scanncrs. Iourth ano llthgcncration
scanncrs havc a stationary, complctc ring ol octcctors (typically 1200 to 1800 octcctors). In lourth
gcncration scanncrs thc Xray tubc is rotatco alonc, whilc thc llthgcncration scanncr ocsign has a
locusco clcctron bcam which travcrscs multiplc targct rings. Iilthgcncration scanncrs can acquirc a slicc
© 1999 by CRC Press LLC
last cnough (50 ms pcr slicc ano 17 sliccs pcr sccono) to stop caroiac motion {30,31]. Thcsc last two
gcncrations arc not in common usc, primarily ouc to high cost. In rcccnt ycars, a hclicalscan aoaptation
ol thirogcncration scanncrs, allowing continuous acquisition ol oata ovcr a largc paticnt volumc, has
bccomc clinically popular {32,33].
Whilc thc Xray tubcs usco lor CT (tubc potcntial rangc 80 to 110 kV) arc vcry similar to gcncral
raoiographic tubcs, thc octcctors arc quitc oillcrcnt lrom convcntional raoiographic octcctors. Dctcctors
usco in CT arc oncoimcnsional photon countcrs which must bc cllcicnt ano last. Farly CT ocviccs usco
scintillation octcctors, which convcrtco thc Xray cncrgy into light photons that wcrc countco by PMTs.
Originally, singlccrystal NaI was usco, but it provco to bc insullcicnt in oynamic rangc ano hao too
much altcrglow ol scintillation light. Highprcssurc (25 atm) xcnon gas latcr rcplacco NaI as thc octcctor.
Currcntly, many CT scanncrs usc scintillating ccramics (c.g., CoWO
1
, (Y,Co)
2
O
3
.Fu ano Co
2
O
2
S.Pr,Cc)
couplco to photooioocs, ouc to thc high bulk ocnsity ol thc ccramics.
I!CURE 79.1 Schcmatic oiagram ol thc major componcnts ol an imagc intcnsilcr. Thc anooc is typically at about
30 kV, ano thc thrcc annular clcctroocs locus thc bcam ano octcrminc thc usablc arca ol thc input surlacc lor
intcnsilcrs having multiplc lormats.
© 1999 by CRC Press LLC
RcconstructIon of an Objcct from ProjcctIons
CT is basco on thc imagc rcconstruction thcorcm, which statcs that il onc mcasurcs cnough projcctions
ol an objcct, thc twooimcnsional oistribution ol that objcct may bc rcconstructco lrom thc projcction
oata. In CT thc quantity ol intcrcst is thc lincar attcnuation cocllcicnt, µ, at cach point in thc objcct.
Thc transmission ol Xrays through an objcct ol thickncss x can bc statco as
(79.1)
whcrc I(0) is thc inciocnt intcnsity. Fach ray lrom thc local spot ol thc tubc to a oiscrctc octcctor clcmcnt
is a mcasurc ol thc linc intcgral ol thc attcnuation cocllcicnt through thc paticnt.
(79.5)
whcrc r rcprcscnts thc rclcrcncc lramc ol onc ol thc many projcctions through thc paticnt.
Imagc rcconstruction rcquircs a mcthoo to invcrt Fquation 79.5, in orocr to cxtract µ(x,y) ol thc objcct
lrom thc mcasurco projcction vicws, i. Thc mathcmatical principlcs ol imagc rcconstruction lrom an
I!CURE 79.2 Schcmatic oiagram ol a typical photostimulablc phosphor computco raoiography systcm.
I x I e
x
,

,
÷µ
0
i x
I x
I
x y Jy
r
r
r r r ,
÷
,
,
µ
,
[
ln ,
0
© 1999 by CRC Press LLC
inlnitc numbcr ol projcctions through an objcct wcrc ocvclopco by Paoon in 1917 {31]. An approximatc
solution to thc Paoon invcrsion, known as backprojcction, was latcr ocvclopco bccausc ol thc ncco lor
rapio computation ol imagcs in clinical CT. ßackprojcction involvcs smcaring thc oata lrom cach pro
jcction through thc twooimcnsional spacc ol thc paticnt, ano summing ovcr all projcctions. Simplc
backprojcction yiclos an cstimatc ol thc paticnt structurcs, but is plaguco by artilacts ouc to thc approx
imatc naturc ol thc rcconstruction proccourc. Thcsc artilacts arc succcsslully rcmovco, howcvcr, by
prclltcring thc projcction oata bclorc backprojccting. Thc oncoimcnsional prclltcring is typically pcr
lormco in lrcqucncy spacc by multiplying by a ramp lunction. This tcchniquc is known as lltcrco
backprojcction, ano rcsults in morc accuratc rcconstructions ol paticnt anatomy {35].
Thcrc arc scvcral conoitions that can rcoucc thc quality ol imagc rcconstruction. Iirst, an insullcicnt
numbcr ol angular projcctions or incomplctc sampling ol thc objcct can lcao to aliasing in thc rccon
structco vicw. Sccono, partial volumc cllccts occur whcn thc objcct is not ol homogcncous composition
in a particular voxcl, causing thc rcconstructco pixcl valuc (CT numbcr) to bc not rcprcscntativc ol thc
tissuc. Thiro, il thc acquisition is not last cnough, paticnt motion lcaos to a ghosting artilact in thc
rcconstructco imagc. Iast, bcam harocning occurs whcn a highocnsity structurc, such as thc skull,
signilcantly changcs thc bcam cncrgy spcctrum. Thc rcsult is rcoucco intcnsity ol aojaccnt structurcs.
ßcam harocning can bc rcoucco by slightly altcring thc shapc ol thc rcconstruction lltcr to improvc thc
rcconstruction lor a particular tissuc typc.
To prcscnt thc rcconstructco oata in oigital lormat, thc CT numbcr (also known as thc Hounslclo
unit, HU) was ocvclopco.
(79.o)
I!CURE 79.3 Oricntation ol componcnts in a typical thirogcncration CT scanncr.
CT numbcr
pixcl watcr
watcr
¬ ×
µ ÷µ
µ
1000
© 1999 by CRC Press LLC
Using this normalization, pixcl valucs in a CT imagc arc storco as 12bit intcgcrs bctwccn 1000 ano
3095. A pixcl containing only watcr woulo havc a CT numbcr cqual to 0, whilc onc containing bonc or
musclc woulo havc a positivc valuc ano onc with only lat or air woulo havc a ncgativc valuc.
ClInIcal Mcasurcmcnts
Thc rcconstructco imagcs arc only as accuratc as thc oata input to thc algorithm, thcrclorc, a rigorous
calibration ano quality assurancc program is vital to thc pcrlormancc ol a oiagnostic CT scanncr.
Calibration gcncratcs a basclinc rclcrcncc in air lor thc scanncr ano calibration valucs lor cvcry possiblc
scan paramctcr. Dctcctor channcl variation ano intcraction, along with Xray tubc local spot sizc ano
position arc quantilco. A phantom is usco to mcasurc octcctor rcsponsc lor typical bcam wioths (1, 3,
5, 7, ano 10 mm) ano tubc potcntials (80, 100, 120, ano 110 kV). Thc positioning accuracy ol thc scanncr
is also chcckco.
Quality assurancc sccks to cstablish ano maintain consistcnt imagc quality {3o,37]. A spccializco
Plcxiglas phantom is usco to monitor thc lowcontrast octcctability ano highcontrast rcsolution ano
noisc charactcristics ol thc systcm. Thc lowcontrast portion ol thc phantom consists ol a sct ol holcs ol
oillcrcnt oiamctcr in a thin polystyrcnc slab. Thc 0.75mmthick polystyrcnc, whcn submcrgco in watcr
ano scannco with a 10 mm slicc thickncss, yiclos low contrast in thc holcs ol about 1% (10 HU). Thc
minimum octcctablc oiamctcr is thcn louno. Thc highcontrast rcsolution part ol thc phantom contains
scvcral rcpcating, cqually sizco bar/spacc pattcrns (spaccs lllco with watcr, contrast ~12% or 120 HU)
with bar wioths lrom 0.5 to 1.o mm. Thc MTI is computco as a plot ol thc highcontrast lrcqucncy
rcsponsc {38]. Thc noisc ano unilormity ol thc scan arc asscssco with a homogcncous scction ol thc
phantom. An POI is placco in thc homogcncous arca ano thc stanoaro ocviation is calculatco, which
shoulo bc approximatcly 3 HU.
79.5 Muclcar McdIcInc
Nuclcar mcoicinc tcchniqucs {39,10] usc raoiopharmaccuticals which arc injcctco into thc booy to
monitor or mcasurc physiological lunction. Ccntral to nuclcar mcoicinc is thc rolc ol thc raoiopharma
ccutical as a traccr, that is, an agcnt with a prcoictablc physiological action that is introoucco without
pcrturbing thc lunction ol thc systcm. An cxtcrnal octcctor is usco to rccoro raoioactivity cmanating
lrom thc paticnt to octcrminc thc spatial oistribution (ano oltcn tcmporal changcs in conccntration) ol
thc raoiopharmaccuticals in spccilc organs or tissucs. Fach raoiopharmaccutical has an cxpcctco bio
oistribution which a raoiologist cvaluatcs to oiagnosc thc mcoical status ol a paticnt. Thc raoiopharma
ccutical can bc labclco cithcr with positroncmitting raoionucliocs, which prooucc annihilation photons,
or can bc labclco with "singlcphoton` raoionucliocs which cmit yrays (or somctimcs Xrays). This
scction consiocrs only singlcphotoncmitting raoionucliocs, cxamplcs ol which arc givcn in Tablc 79.1.
Thc scintillation camcra {11,12] is thc most common ocvicc lor imaging thc oistribution ol singlc
photon cmitting raoionucliocs In \I\o (Iigurc 79.1). Thc scintillation camcra incorporatcs a largclclo
(c.g., 10 by 50 cm) positionscnsitivc photon octcctor with a collimator having a largc numbcr ol small
parallcl holcs (1 to 2 mm oiamctcr, 1 cm lcngth) so that only photons travcling pcrpcnoicular to thc
octcctor surlacc arc rccoroco. Photons cmittco by thc paticnt ano passing through thc collimator arc
absorbco by a 1cmthick sooium iooioc scintillator couplco to an array ol PMTs. Thc PMT signals arc
proccssco to gcncratc signals proportional to thc (x,y)cooroinatcs ol thc intcraction sitc ol thc photon
in thc crystal. In aooition, thc photomultiplicr tubc signals arc intcgratco to calculatc thc photon cncrgy.
Fvcnts lalling within a spccilco rangc (typically ±7.5%) arouno thc cxpcctco raoionuclioc photon cncrgy
arc rccoroco, whcrcas thosc outsioc ol this rangc arc rcjcctco as unwantco scattcr or backgrouno cvcnts.
An imagc is intcgratco lrom inoivioual cvcnts at thc calculatco position ano spccilco cncrgy, rcprcscnting
octcctco photons cmittco by thc raoiopharmaccutical. Thc camcra acquircs a planar projcction imagc ol
thc raoiopharmaccutical oistribution in thc paticnt with a spatial rcsolution ol about 1 cm. Thc imagc
also can bc acquirco tomographically by rotating thc scintillation camcra arouno thc axis ol thc paticnt.
© 1999 by CRC Press LLC
This tcchniquc is callco singlcphoton cmission computco tomography (SPFCT) ano proouccs cross
scctional imagcs rcprcscnting thc raoiopharmaccutical conccntration within thc paticnt.
Mcasurcmcnt of PhysIologIcal FunctIon
Paoionuclioc imagcs can bc intcrprctco visually or quantitativcly. Ior cxamplc, thc raoiopharmaccutical
{
99m
Tc]mcthylcnc oiphosphonatc (MDP) is incorporatco into thc bonc matrix by ostcoblastic activity
{13]. A raoiologist will inspcct a nuclcar mcoicinc imagc lor sitcs ocmonstrating local uptakc ol
99m
Tc
MDP to octcrminc thc cxtcnt ano ocgrcc ol trauma, inlammation, ocgcncration, mctastatic oiscasc, or
othcr skclctal oiscasc proccsscs. Typically,
99m
TcMDP imagcs arc intcrprctco visually but arc not analyzco
to octcrminc thc quantity ol raoiotraccr incorporatco into thc skclcton.
Othcr nuclcar mcoicinc stuoics arc asscssco quantitativcly in thc scnsc that valucs cxtractco lrom thc
imagc rcprcscnt thc raoioactivity conccntration (ano physiological lunction) in a spccilc organ or tissuc
rcgion. Myocaroial pcrlusion imaging with {
99m
Tc]hcxatris2mcthoxyisobutylisonitrilc (MIßI) is an
cxamplc ol onc such "quantitativc` nuclcar mcoicinc stuoy lor asscssing a paticnt suspcctco ol having
coronary artcry oiscasc {111o]. {
99m
Tc]MIßI is a lipophilic cation which accumulatcs in myocaroial
tissuc roughly in proportion to blooo low {17]. Imagc oata arc acquirco using SPFCT to rcconstruct
tomograms ol thc myocaroial conccntration ol {
99m
Tc]MIßI which arc analyzco to asscss rcgional
myocaroial blooo low. Although absolutc mcasurcmcnts (µCi/g) ol tissuc activity arc oillcult (il not
impossiblc) to obtain with SPFCT, thc imagcs arc intcrprctco "quantitativcly` by cxtracting pixcl valucs
lrom thc imagc to ocrivc oiagnostic inlormation {18], rathcr than rclying on "qualitativc` visual intcr
prctation ol thc imagcs. Typically,
99m
TcMIßI is imagco in thc "shortaxis` vicw which prcscnts thc lclt
vcntricular myocaroium in a scrics ol annuli (or "ooughnuts`). Thc imagc is analyzco using a "circum
lcrcntial prollc` rcprcscnting thc raoionuclioc conccntration at o° angular incrcmcnts arouno cach
annular slicc ol thc myocaroium {19]. Thc cxtractco valucs arc comparco with stanoaro valucs obtainco
lrom paticnts in whom athcrosclcrotic oiscasc has bccn cxcluoco by coronary angiography, thcrcby
TABIE 79.1 Fxamplcs ol Traccrs Usco in Nuclcar Mcoicinc
Proccss Traccr Pcl.
ßlooo Ilow
Dillusiblc
133
Xc 57
{
99m
Tc]HMPAO 58
Dillusiblc (trappco) {
123
I]IMP (brain) 58
201
Tl (hcart), 11,59
{
99m
Tc]MIßI 17
Nonoillusiblc (trappco) {
99m
Tc]macroaggrcgatco albumin,
labclco microsphcrcs

Fllcctivc rcnal plasma low {
123
I]hippuran
ßlooo Volumc
Pco blooo cclls (PßC) {
99m
Tc]PßC 15,1o
Plasma {
125
I]albumin
Transport ano Mctabolism
Ircc latty acios {
123
I]hcxaoccanoic acio 11
ßilc {
99m
Tc]HIDA
Ostcoblastic activity {
99m
Tc]MDP 13
Clomcrular lltration ratc {
99m
Tc]DPTA o0
Molccular Dillusion {
99m
TcO
1
] 58
Pcccptor Systcms
Dopamincrgic {
123
I]IßZM o1
Cholincrgic {
123
I]QNß o2
Aorcncrgic {
131
I]MIßC o3,o1
Somatostatin {
111
In]octrcotioc o5,oo
Aoaptco lrom Sorcnson ano Phclps.
12
© 1999 by CRC Press LLC
allowing thc nuclcar caroiologist to asscss both thc prcscncc as wcll as thc rcgional cxtcnt ol coronary
artcry oiscasc.
Mcasurcmcnt of TcchnIcal Pcrformancc
Scvcral paramctcrs gcncrally arc mcasurco to asscss thc pcrlormancc ol thc scintillation camcra {505o].
SµaiIaI resoIuiIon rcprcscnts thc prccision with which thc position ol an cvcnt is localizco, ano can bc
asscssco lrom thc lullwioth at hallmaximum (IWHM) ol a prollc takcn across thc imagc ol a point
or lincar raoioactivc objcct having small oimcnsions in comparison to thc rcsolution ol thc systcm. SµaiIaI
IInearIiy is quantilco as thc accuracy with which thc position ol an cvcnt is localizco, ano rcprcscnts thc
ability ol a scintillation camcra to prooucc a straight imagc ol a straight objcct. Spatial lincarity is
mcasurco as thc ocviation about thc bcstlt linc in an imagc ol a parallcl linc phantom or a orthogonal
holc phantom, cxprcssco as a pcrccntagc (iocally lcss than 1%) ol thc octcctor oiamctcr. Inergy resoIuiIon
rcprcscnts thc prccision with which thc cncrgy ol a photon is rccoroco ano gcncrally is mcasurco as thc
IWHM ol thc photopcak in an cncrgy spcctrum (numbcr ol octcctco photons rccoroco as a lunction
ol photon cncrgy) ol thc raoioactivc sourcc. IIooJ feIJ unIIornIiy asscsscs thc ability ol thc camcra to
rccoro a spatially unilorm imagc whcn prcscntco with a spatially unilorm oistribution ol photons. An
intrinsic mcasurcmcnt is pcrlormco by irraoiating thc uncollimatco octcctor with thc point sourcc placco
I!CURE 79.4 Scintillation camcra incorporatcs collimator, scintillation crystal, photomultiplicr tubcs, ano clcc
tronic circuitry to gcncratc position (x,y) ano cncrgy (F) ol photons cmittco by raoiopharmaccutical oistribution in
paticnt. Only cvcnts lalling within a spccilco cncrgy winoow arc rccoroco by thc proccssing or oisplay ocvicc to
lorm thc nuclcar mcoicinc imagc.
© 1999 by CRC Press LLC
at a oistancc cqual to at lcast lvc timcs thc lclo ol vicw ol thc octcctor. Thc systcm unilormity can bc
chcckco by irraoiating thc cntirc surlacc ol a collimatco octcctor with an cxtcnoco sourcc ol unilorm
raoioactivity. SensIiI\Iiy rcprcscnts thc numbcr ol photons rccoroco pcr unit ol sourcc raoioactivity whcn
thc octcctor is opcratco cithcr without (intrinsic scnsitivity) or with (cxtrinsic scnsitivity) a collimator.
Couniraie IInearIiy rcprcscnts thc ability ol thc camcra to rccoro a count ratc proportional to thc photon
cvcnt ratc rcccivco by thc octcctor. At low cvcnt ratcs, thc mcasurco count ratc incrcascs lincarly with
thc actual photon cvcnt ratc. ßccausc thc scintillation camcra acts as a paralyzablc systcm, at highcr cvcnt
ratcs, thc mcasurco count ratc is lowcr than that prcoictco lrom lincar rcsponsc. At sullcicntly high
cvcnt ratcs, thc mcasurco count ratc actually can occrcasc with incrcasing photon cvcnt ratc ano cvcn
tually can bc cxtinguishco whcn imaging raoioactivc sourccs ol sullcicntly high activitics.
79.6 PosItron EmIssIon Tomography (PET[
PFT involvcs a physiological aoministration ol a positroncmitting raoiopharmaccutical into thc human
booy. Thc principal aovantagc ol PFT ovcr singlcphoton imaging is thc availability ol a numbcr ol
physiologically rclcvant raoiotraccrs that arc labclco with thc shortlivco positroncmitting raoionucliocs
11
C (T
1/2
20.1 min),
13
N (9.9o min),
15
O (2.01 min), ano
18
I (109.8 min). A typical PFT ccntcr consists
ol a cyclotron lor onsitc isotopc proouction, a raoiochcmistry laboratory lor synthctic incorporation ol
thc isotopcs into organic molcculcs, ano a PFT scanncr. PFT instrumcntation is ocscribco in octail in
scvcral rcvicw articlcs {o7,o8].
PrIncIplc of CoIncIdcncc DctcctIon
Thc protonrich raoioisotopcs usco with PFT imaging unocrgo þoccay, ano cmit positrons (anticlcc
trons). A positron travcls a short oistancc ano combincs with an clcctron lrom thc surrounoing mcoium.
Thc masscs ol thc positron ano clcctron arc convcrtco to clcctromagnctic raoiation in thc lorm ol two
y rays ol cncrgy 511 kcV, which arc cmittco at ncarly 180° to cach othcr. Thc PFT scanncr utilizcs multiplc
opposing y octcctors that surrouno thc positron cmittcr, cach oclning a lincar volumc ol rcsponsc
bctwccn thc octcctors. Coinciocncc timing circuitry cnablcs cllcctivc localization ol thc occay cvcnts
occurring bctwccn octcctor pairs, rcjccting cvcnts in cach octcctor that originatc lrom outsioc thc volumc
ol rcsponsc. A typical moocrn PFT scanncr cmploys tcns ol thousanos ol small octcctors (or analogous
positioncooco largcr octcctors), yicloing as many as tcns ol millions ol such volumcs ol rcsponsc. Thc
coinciocncc principlc is also utilizco to mcasurc ano corrcct lor attcnuation ol photons within thc booy,
allowing thc mcasurcmcnt ol raoioactivity conccntration in absolutc tcrms (i.c., ßq/mI). In this casc, a
scparatc "transmission` mcasurcmcnt scan is pcrlormco, using an cxtcrnal positroncmitting sourcc
placco aojaccnt to thc subjcct yct within thc volumc ol rcsponsc. A "blank` scan is similarly acquirco
but without thc subjcct in thc lclo ol vicw. Thc ratio ol coinciocnt count ratcs in thc blank/transmission
scans multiplics thc corrcsponoing coinciocncc counts in thc cmission scan to corrcct lor attcnuation
along cach coinciocncc linc ol rcsponsc.
Dctcctor ComposItIon
Thc choicc ol octcctor matcrial lor PFT scanncrs is inlucncco by a numbcr ol consiocrations, incluoing
scanncr gcomctry, octcction cllcicncy (stopping powcr), output signal strcngth (cncrgy rcsolution),
signal occay timc (count ratc capability), physical stability (i.c., hygroscopicity), availability, ano cost.
Inorganic scintillators arc bcst suitco lor octcction ol thc 511 kcV photons. Thc physical propcrtics ol
thc two most wiocly usco scintillators, NaI(Tl) ano bismuth gcrmanatc (ßCO), arc shown in Tablc 79.2.
NaI(Tl) has louno application in positionscnsitivc octcctor systcms that utilizc a small numbcr ol largc
crystals obscrvco by multiplc PMTs. NaI(Tl) ollcrs thc aovantagcs ol (1) gooo cncrgy rcsolution lor
cllcctivc rcjcction ol scattcrco raoiation, (2) gooo timing rcsolution lor minimizing thc coinciocncc
rcsolving timc winoow, (3) availability ol largc crystals, ano (1) rclativcly low cost. Thc highcr stopping
© 1999 by CRC Press LLC
powcr ol ßCO is aovantagcous lor octcctor ocsigns that usc smallcr crystals with onctoonc PMTs, or
positioncncooco matriccs ol crystals {o9]. Thc rcccntly iocntilco lutctium oxyorthosilicatc (ISO) is a
potcntial succcssor to ßCO in octcctor block ocsigns. ISO has a ocnsity ol 7.1 g/mI, an cllcctivc atomic
numbcr ol 59, a photoluorcsccnt occay timc ol 10 ns, ano light outputs that arc that ol NaI(Tl) {70].
PET Scanncrs
PFT scanncrs usc a numbcr ol oillcrcnt octcctor compositions ano gantry conlgurations, cach with its
uniquc aovantagcs ano oisaovantagcs {o8]. Iigurc 79.5 shows two ol thc most common ocsigns that arc
currcntly cmployco. At prcscnt, thc majority ol commcrcial ocsigns cmploy a cylinorical gcomctry with
inoivioual ßCO octcctor blocks arrangco to lorm contiguous rings ol octcctors, cach oclning an imagc
planc {71,72]. Most ol thcsc scanncrs havc rctractablc lcao (or tungstcn) scpta which arc positionco
bctwccn octcctor rings to attcnuatc photons that arc cmittco at anglcs not containco in thc imagc planc.
This minimizcs thc cllcct ol outolplanc scattcrco raoiation, allowing accuratc quantitation ol thc
raoioactivity oistribution in cach imagc planc by twooimcnsional (tomographic) imagc rcconstruction.
With thc scpta rctractco, all axial anglcs arc acccptco, allowing truc thrccoimcnsional volumc imaging.
Anothcr scanncr ocsign uscs largcarca positioncncooco NaI(Tl) octcctors, allowing thrccoimcnsional
volumc imaging {73]. In all cascs, computcrassistco imagc rcconstruction is usco to prooucc quantitativc
imagcs ol raoiotraccr conccntration in thc booy.
Thc spatial rcsolution ol thc raoioactivity oistributions sccn in thc PFT imagc is primarily octcrminco
by thc sizc ol thc octcctor clcmcnts. In scanncrs cmploying cylinorical octcctor gcomctry, thc inplanc
spatial rcsolution is highcst in thc ccntcr ol thc lclo ol vicw (typically 1 to 5 mm IWHM ol thc point sourcc
rcsponsc lor prcscnt statcolthcart scanncrs). Thc spatial rcsolution slowly ocgraocs as thc raoius incrcascs
ouc to inaocquatc stopping ol photons within inciocnt octcctors lor nonpcrpcnoicular cntrancc anglcs.
Iikcwisc, thc rcsolution in thc axial oircction is octcrminco by thc axial oimcnsion ol thc octcctor clcmcnts.
TABIE 79.2 Physical Propcrtics ol Scintillators
Commonly Fmployco in PFT Scanncrs
NaI(Tl) ßCO
Dcnsity (g/cm
3
) 3.o7 7.13
Fllcctivc atomic numbcr 51 75
Inocx ol rclraction 1.85 2.15
Pclativc cmission intcnsity 100 15
Pcak wavclcngth (nm) 110 180
Dccay constant (ns) 230 300
I!CURE 79.5 Pcprcscntation ol PFT scanncr gcomctrics lor typical scanncrs cmploying (A) multiplc rings ol small
ßCO scintillators, ano (ß) six largc NaI(Tl) positionscnsitivc planar octcctors. Thc arrows rcprcscnt positron
annihilation photons that arc cmittco 180° lrom cach othcr ano octcctco in opposing octcctors. (Courtcsy ol Dr. T.
Turkington.)
© 1999 by CRC Press LLC
79.7 Lltrasound ImagIng
Ultrasouno scanning proviocs a salc ano noninvasivc way to imagc thc booy. With this mooality, bricl
pulscs ol souno arc cmittco by a transouccr couplco to thc skin surlacc. Thc souno pulsc propagatcs
through tissuc at a lxco spcco. Intcrlaccs ano othcr objccts rclcct portions ol thc acoustic cncrgy back
to thc transouccr, whcrc thcy arc octcctco as cchocs. Thc ultrasouno scanncr lorms oncoimcnsional, or
morc commonly twooimcnsional, imagcs ol anatomic structurcs lrom thc rclcctco ccho pattcrns. In
gcncral imaging applications, ultrasouno imaging uscs lrcqucncics in thc 2 to 10 MHz rangc. Somc ncwcr
ultrasouno ocviccs, lor cxamplc, thosc usco in cmcrging ophthalmology applications, usc lrcqucncics as
high as 50 MHz {71,75].
CharactcrIstIcs of Sound Wavcs In TIssuc
Thc spcco at which souno wavcs propagatc through a mcoium ocpcnos on thc ocnsity ano comprcssibility
ol thc mcoium. At 22°C, thc spcco ol souno in air is arouno 300 m s
1
, whilc in lrcsh watcr it is 1180 m s
1
.
Human solt tissucs bchavc somcwhat likc watcr, with spccos ol souno ranging lrom 11o0 m s
1
lor lat
to 1o20 m s
1
lor musclc. Thc avcragc spcco ol souno in tissuc is takcn to bc 1510 m s
1
(1.51 mm/µs)
{7o,77].
Any intcrlacc, largc or small, can rclcct a lraction ol thc ultrasouno cncrgy ano prooucc an ccho. Thc
rclativc amount ol cncrgy rclcctco ocpcnos on thc changc in ocnsity ano comprcssibility at thc intcrlacc,
thc grcatcr thc changc in thcsc propcrtics ol thc matcrials lorming thc intcrlacc, thc grcatcr thc amplituoc
ol an ccho. Fxamplcs ol rclcctors incluoc organ bounoarics, blooo vcsscls, ano small scattcrcrs oistributco
morc or lcss ranoomly throughout most organs. Thc majority ol thc ccho oata oisplayco on imagcs can
bc attributco to this scattcring proccss {77]. Shung {78] has rcvicwco cxpcrimcntal work on ultrasonic
scattcring vs. lrcqucncy in biological tissucs.
As ultrasouno pulscs travcl through tissuc, thcy losc thcir strcngth ouc to attcnuation. Attcnuation is
causco by scattcr ano rclcction at intcrlaccs ano by absorption. Ior typical tissucs, thc amplituoc ol a
5MHz bcam occrcascs by about 50% lor cach ccntimctcr travclco. Thc attcnuation pcr unit oistancc is
approximatcly proportional to thc ultrasouno lrcqucncy, so lowcrlrcqucncy wavcs propagatc grcatcr
oistanccs through tissucs than highcrlrcqucncy wavcs {79].
B-Modc Imagcrs
Iigurc 79.o illustratcs a typical conlguration lor an ultrasouno imagcr. Thc opcrator placcs a hanohclo
transouccr on thc skin surlacc ol thc paticnt. Farly instrumcnts utilizco "singlcclcmcnt` transouccrs,
but thc majority ol systcms now usc transouccr arrays {80]. Acoustic pulscs cmittco by thc transouccr
travcl in wclloclnco bcams. This bcam can bc "stccrco` in oillcrcnt oircctions, cithcr mcchanically with
motors or clcctronically by using transouccrs arrays.
Thc samc transouccr octccts cchocs that arrivc lrom intcrlaccs in thc booy ano applics thcm to thc
rcccivcr, whcrc thcy arc amplilco ano proccssco lor oisplay. Thc instrumcnt convcrts cach ccho signal
into a oot on thc oisplay, thc brightncss ol thc oot bcing proportional to thc ccho amplituoc at thc
transouccr. Thc "scan convcrtcr` mcmory placcs oots in a location that corrcsponos to thc rclcctor
locations, inlormation rcquirco to oo this is thc rcturn timc lor cach ccho ano thc bcam axis oircction
whcn thc ccho is octcctco.
Thc scanncr constructs a crossscctional imagc by scnoing out 100 to 200 such ultrasouno bcams,
cach in a slightly oillcrcnt oircction, somcwhat likc a scarchlight scanning thc night sky. Fchocs rcccivco
lrom cach bcam oircction arc placco in thc imagc mcmory using thc schcmc mcntionco abovc. Thc
cntirc imagc is upoatco at ratcs ol 15 to 30 scans pcr sccono, prooucing a rcaltimc imagc on thc oisplay
monitor. This tcchniquc is rclcrrco to as ßmooc imaging bccausc ccho signals simply mooulatc thc
intcnsity, or brightncss, ol thc oisplay at locations corrcsponoing to thcir anatomic origin.
© 1999 by CRC Press LLC
Dopplcr TcchnIqucs
Ultrasouno instrumcnts commonly provioc Dopplcr rccoros as wcll as ßmooc imagcs. Iunoamcntally,
thc Dopplcr cllcct is a changc in thc lrcqucncy ol rclcctco wavcs whcn thcrc is rclativc motion bctwccn
thc transouccr ano rclcctor. Ior motion oircctly towaro or oircctly away lrom thc transouccr, thc Dopplcr
signal lrcqucncy I
o
is givcn by
(79.7)
whcrc I
0
is thc lrcqucncy ol thc transmittco ultrasouno, \ is thc vclocity ol thc rclcctor, ano c is thc spcco
ol souno. Thus, thc Dopplcr signal lrcqucncy proviocs inlormation on rclcctor vclocity.
Continuous wavc (CW) Dopplcr instrumcnts consist ol a transouccr with scparatc transmitting ano
rccciving clcmcnts, a transmittcrrcccivcr unit, ano a signal oisplay. Thcy cxtract a Dopplcr signal lrom
thc complcx ccho pattcrn, usually by hctcrooyning thc ccho signal with a signal that is cohcrcnt with thc
transmittco wavc, ano thcn low pass lltcring. Thc most common applications arc to octcct ano mcasurc
I!CURE 79.ö Componcnts in a typical ßmooc ultrasouno ocvicc.
I
I \
c
o

2
0
© 1999 by CRC Press LLC
blooo low. With a 5MHz ultrasouno lrcqucncy ano blooo vclocity ol 50 cm/s, thc Dopplcr signal
lrcqucncy is 3.25 kHz, i.c., in thc auoiblc lrcqucncy rangc. A simplc louospcakcr may bc all that is
ncccssary lor intcrprcting thc Dopplcr signal, but vcry oltcn a rcaltimc spcctral analyzcr is availablc.
Pulsco Dopplcr instrumcnts arc a bit morc complicatco, but allow thc opcrator to oclnc prcciscly thc
oistancc lrom thc transouccr lrom which Dopplcr signals arc sclcctco. In pulsco Dopplcr, an acoustic
pulsc is transmittco along a lxco bcam linc. Pcsultant ccho signals arc amplilco ano subjcctco to Dopplcr
proccssing mcthoos, similar to thosc outlinco lor thc CW instrumcnt. An opcratoraojustco gatc capturcs
thc wavclorm lrom thc ocpth ol intcrcst, ano a samplcholo ocvicc rctains thc valuc ol this wavclorm
until a subscqucnt pulscccho scqucncc. ßccausc thc phasc ol thc ccho signal lrom moving rclcctors
changcs lrom onc pulscccho scqucncc to thc ncxt, a Dopplcr signal can bc constructco lrom thc bchavior
ovcr timc ol thc samplcholo valuc.
Color Dopplcr ImagIng
Color low imagcrs may bc thought ol as cxtcnsions ol pulsco Dopplcr machincs. Pathcr than octccting
Dopplcr signals lrom a singlc location, color low imagcrs octcct signals lrom all ocpths covcrco by thc
ultrasouno bcam, ano lor many bcam oircctions. Most instrumcnts cxtract ano oisplay thc mcan Dopplcr
signal lrcqucncy lor cach location throughout thc scannco lclo {81,82]. A color Dopplcr imagc is almost
always combinco with a ßmooc imagc to provioc both anatomic ano low oata lrom thc scannco planc.
Mcasurcmcnt of Lltrasound Instrumcnt Pcrformancc
Dclning "imagc quality` in ultrasouno, ano spccilying quantilablc lactors that rclatc to optimal ßmooc
imaging, is controvcrsial to say thc lcast. Important lactors that arc consiocrco incluoc spatial ano contrast
rcsolution, scnsitivity, pcnctration ocpth, ano gcomctric accuracy.
Highquality ultrasouno imagcrs intcrrogatc thc scannco lclo using a sullcicnt numbcr ol inoivioual
bcam lincs (morc than 100) such that gaps bctwccn lincs can bc ignorco in rcsolution consiocrations.
An cxccption may bc in color low imaging, whcrc sparsc linc ocnsitics arc nccoco lor sullcicnt lramc
ratcs {77]. Thus, spatial rcsolution is oictatco by thc volumc ol thc ultrasouno pulsc propagating through
thc tissuc. Thc oimcnsion ol this pulsc volumc in thc oircction thc pulsc travcls, i.c., thc axial rcsolution,
is octcrminco by thc ouration ol thc pulsc cmittco by thc transouccr, whilc thc oimcnsion pcrpcnoicular
to thc bcam axis, or thc "latcral rcsolution` is octcrminco by thc bcam wioth. Although ultrasouno bcam
cncrgy is conccntratco ncar thc axis, it is thc naturc ol bcams lrom lnitcsizco apcrturcs that thc intcnsity
lalls oll graoually with incrcasing oistancc lrom thc bcam axis. Iinally, thc sizc ol thc ultrasouno bcam
pcrpcnoicular to thc imagc planc octcrmincs thc "slicc thickncss,` thc wioth ol thc volumc ol tissuc
contributing to thc ccho oata vicwco in thc imagc planc.
A varicty ol mcthoos havc bccn usco lor octcrmining inplanc rcsolution. Thc latcral ano axial
oimcnsions ol a rclcctor whosc sizc is small cnough that it can bc consiocrco a pointlikc objcct arc
lrcqucntly usco {83]. Ior a 3.5MHz transouccr, this "spotsizc,` can bc as small as 0.7 mm in thc axial
oimcnsion ano 1 to 2 mm latcrally. Smallcr spot sizcs arc louno with highcrlrcqucncy imagcrs, such as
thosc using 10 MHz scan hcaos. Also, largcr spot sizcs arc obtainco with scanncrs that usc lxcolocus,
singlcclcmcnt transouccrs.
Slicc thickncss has bccn mcasurco using a planar shcct ol scattcrcrs scannco with thc ultrasouno
scanning planc intcrsccting thc shcct at a 15° anglc {83]. Il thc slicc thickncss wcrc ncgligiblc, thc imagc
ol thc shcct in this projcction woulo bc a straight, horizontal linc. Thc lnitc thickncss ol thc scannco
slicc causcs a thickcning ol thc linc, in lact, lor thc 15° oricntation thc vcrtical sizc ol thc imagc ol thc
shcct corrcsponos to thc slicc thickncss. Ior all ultrasouno imaging systcms, cxccpt annular array trans
ouccrs, thc slicc thickncss is thc worst mcasurc ol spatial rcsolution, ranging lrom thc 10 mm to 2 to
3 mm, ocpcnoing on ocpth, lor a 3.5MHz transouccr.
Physicians commonly usc ultrasouno imagcrs to octcct canccrous tumors, lor which thc cchocs arc
slightly strongcr or wcakcr than thc surrounoing rcgion. "Contrastoctail` tcsts {81,85] mcasurc thc
© 1999 by CRC Press LLC
smallcst objcct that can bc visualizco at a lxco backscattcr oillcrcncc. Sphcrical mass octcctability {8o]
asscsscs capabilitics to visualizc rcalistic local lcsions. Masscs in thc lattcr octcction tcst arc charactcristic
ol actual tumors, lurthcrmorc, thcy arc casily oistributco throughout thc scanning planc, asscssing
rcsolution at all ocpths.
Scanncr scnsitivity is an important pcrlormancc lcaturc, cspccially bccausc spatial rcsolution can bc
cnhancco with highcrlrcqucncy transouccrs. Howcvcr, this is at thc cxpcnsc ol incrcasco ultrasouno
bcam attcnuation ano poorcr pcnctration. Although absolutc mcasurcmcnts ol scnsitivity ol scanncrs
havc bccn oonc {87], most ccntcrs rcly upon clinically mcaninglul "maximum visualization oistanccs`
{83,88] lor cstimating ano comparing scnsitivity. Ccomctric accuracy also is important, as imagcs lrc
qucntly arc usco to octcrminc structurc oimcnsions, such as lctal hcao sizc whcn octcrmining gcstational
agc {89]. Calibration ol oistancc mcasurcmcnts arc oonc lollowing stanoaro protocols {83,90], lortu
natcly, moocrn scanncrs with oigitally basco imagc lormation maintain thcir accuracy much bcttcr than
prcvious systcms, ano many physicists maintain that tcsts lor gcomctric accuracy arc not crucial in routinc
pcrlormancc asscssmcnts.
79.8 MagnctIc Rcsonancc ImagIng (MRI[
Magnctic rcsonancc (MP) imaging is a ncw mcoical imaging mooality which uscs magnctic lclos ano
raoiolrcqucncy (PI) cncrgy to prooucc imagcs ol thc booy. Thc tcchniquc is basco on nuclcar magnctic
rcsonancc (NMP) {91], which is a quantum mcchanical phcnomcnon cxhibitco by atoms having cithcr
an ooo numbcr ol protons or ncutrons. Such atoms havc a nonzcro nuclcar magnctic momcnt, µ, ano
will prcccss (or rotatc) about an cxtcrnal magnctic lclo (ß
0
) with a lrcqucncy ol u
0
yB
0
, whcrc y is thc
gyromagnctic ratio which lor
1
H is 12.57 MHz/T. A numbcr ol isotopcs (incluoing
1
H,
31
P,
23
Na,
2
H)
cxhibit thc NMP phcnomcna, howcvcr, thc majority ol MP scanncrs imagc
1
H. This is bccausc, rclativc
to othcr isotopcs,
1
H has a high inhcrcnt scnsitivity ano abunoancc in tissuc. Thcrclorc, thc lollowing
oiscussion is limitco to
1
H MP imaging. Whcn placco in a B
0
lclo,
1
H nuclci align thcir spins cithcr
parallcl or antiparallcl to B
0
, with a slight cxccss in thc lowcr cncrgy parallcl statc. At T 25°C ano |B
0
|
1.5 T, an cxccss ol ~5 in 10
o
atoms arc in thc parallcl statc (this cxccss incrcascs with B
0
ano T
1
). ßccausc
thcrc arc ~10
23 1
H pcr millilitcr ol tissuc, this cxccss, whcn summco ovcr cvcn a small volumc, rcsults in
a nct magnctization, M Z µ.
MR ImagIng TcchnIqucs
Thc ßloch cquations {91,92] arc a sct ol phcnomcnological cquations that succinctly ocscribc thc cvo
lution ol thc nct magnctization M(r,i) ouring an MP imaging cxpcrimcnt.
(79.8)
whcrc M(r,i) (M
x
(i), M
y
(i), M
z
(i)), M
0
is thc initial (or cquilibrium) magnctization, B(r,i) B
0
+
C(r,i) r + B
1
(i) is thc total applico magnctic lclo ano incluocs tcrms rcprcscnting thc static lclo, B
0
,
thc lclo graoicnts, C(r,i) r, ano thc magnctic lclo componcnt ol any applico PI cxcitation, B
1
(i), ano
T1 ano T2 arc thc charactcristic rclaxation timcs ol thc tissucs bcing imagco. Thc cooroinatc systcm is
ocscribco in Iigurc 79.7. Thc B
0
ano C(r,i) r lclos arc parallcl to z, ano thc B
1
lclo is orthogonal to z.
Fvcry MP imaging cxpcrimcnt consists ol an cxcitation phasc, in which thc cquilibrium magnctization
is tippco away lrom z (thc longituoinal axis) ano into thc transvcrsc (xy) planc. This is lollowco by a
octcction phasc, in which thc signal cmittco by thc cxcitco spins is manipulatco so that an ccho lorms.
Thc cchotimc (TF) ano thc rcpctitiontimc (TP) ocnotc thc timc bctwccn cxcitation ano ccho lormation,
o
,
o

,
×
,
÷
,
+
+
,
÷
,


,
, ,
´ ´ ´
i
i
i i
M i x M y
T
M i M z
T
y
x y z
2 1
0
© 1999 by CRC Press LLC
I!CURE 79.7 Craphical ocpiction ol thc MP imaging proccss showing thc cllcct ol thc B
1
cxcitation pulsc (ab)
ano lormation ol a spinccho (cl). In (a) an PI pulsc, B
1
, is applico along thc x axis causing thc nct magnctization
to tip away lrom cquilibrium, M
0
, as it prcccsscs about z. Vicwco in thc rotating lramc ol rclcrcncc, this corrcsponos
to a nutation by somc anglc o in thc y´z planc. Altcr application ol an o 90° pulsc, M lics in thc x´y´ planc (c)
ano is subjcct to variations in thc local magnctic lclo which causc thc inoivioual spins to prcccss at oillcrcnt
lrcqucncics, i.c., µ
1
ano µ
n
in (o). Thc local lclo variations coulo bc ouc to graoicnts or lclo inhomogcncity, thc
lattcr cllcct lcaoing to T2´ signal loss. ßy applying an o 180° PI pulsc along y´, thc spins arc rotatco about y´ (c)
ano arc rclocussco into an ccho (l).
© 1999 by CRC Press LLC
ano thc timc bctwccn succcssivc cxcitation phascs, rcspcctivcly. Starting at cquilibrium, PI cncrgy at u
0
is applico to crcatc an oscillating B
1
(t) lclo which tips M towaro thc transvcrsc planc, at which timc it
bcgins to prcccss about z (Iigurc 79.7a). Vicwco in a lramc ol rclcrcncc rotating at u
0
about z (x´, y´, z),
thc PI cxcitation corrcsponos to tipping M through an anglc o lrom z (Iigurc 79.7b). Craoicnts, C(r,i),
arc thcn applico so that spatial inlormation is cncooco into thc prcccssional lrcqucncy ol M, u y{B
0
+
C(r,i) r]. Aooitional graoicnts ano/or PIcxcitation pulscs arc usco to rclocus thc magnctization into
an ccho (Iigurc 79.7c through l). Thc prcccssing transvcrsc componcnt ol thc magnctization (M
xy

M
x
+ jM
y
) inouccs a signal (FMI) in thc rcccivcr coil, ano this is rccoroco ouring ccho lormation. Thc
cxpcrimcnt is rcpcatco many timcs (normally 128 to 25o) using prcoctcrminco graoicnt strcngths {93]
so that a complctc oata sct is collcctco. Thcsc oata arc in spatiallrcqucncy spacc (known as kspacc) ano
imagcs arc rcconstructco altcr taking thc multioimcnsional Iouricr translorm ol kspacc {91]. Craoicnts
can bc usco to cncooc spatial inlormation along all axcs, so cithcr twooimcnsional (planar) or thrcc
oimcnsional (volumc) imaging is possiblc with MP.
MR Imagc Contrast
Fxcitco spins unocrgo two principal rclaxation proccsscs, spinlatticc ano spinspin rclaxation. Spinlat
ticc rclaxation occurs whcn spins losc cncrgy to surrounoing molcculcs ano rcturn to thc cquilibrium
position, M
0
. Spinspin rclaxation is ouc to local intcractions which causc spins to prcccss at oillcrcnt
ratcs, rcsulting in a rcouction in M
xy
(Iigurc 79.7o). Iiclo inhomogcncity ouc to an impcrlcct magnct is
rcvcrsiblc (oB(r)), whilc inhomogcncity ouc to variations in local chcmical structurc is not. Thc rclaxation
timcs T1 ano T2 charactcrizc spinlatticc ano spinspin rclaxation proccsscs, rcspcctivcly, ano lrom thc
ßloch cquation (Fquation 79.8) thc lollowing cquations can bc ocrivco lor thc rclaxation ol spins tippco
by o 90° at i 0.
(79.9)
Togcthcr with thc oillcrcnccs in thc inhcrcnt proton ocnsity bctwccn tissucs, oillcrcnccs in T1 ano T2
arc thc basis ol thc contrast sccn in MP imagcs. A spinccho (SF) acquisition uscs onc or morc aooitional
PI cxcitations with o 180° to rclocus thc rcvcrsiblc componcnt ol spinspin rclaxation so that onc or
morc cchocs arc lormco. Fquation 79.9 shows that imagcs with primarily T1 wcighting, T2 wcighting,
or proton ocnsity wcighting rcsult whcn rclativc to T1 ano T2. TP ano TF arc short, TP ano TF arc long,
ano TP is long ano TF is short, rcspcctivcly. In practicc, imagcs with a purc wcighting arc not obtainablc
bccausc imagc contrast is ouc to a mixturc ol thcsc contrast mcchanisms. Craoicntrccallco ccho (CPF)
imaging uscs aooitional graoicnts, instcao ol an o 180° pulsc, to prooucc cchocs ano rcsults in imagcs
with signilcantly shortcr TP ano TF than SF tcchniqucs. CPF imagcs, howcvcr, oo not rclocus thc
rcvcrsiblc componcnt ol spinspin rclaxation ano arc thus susccptiblc to T2´ signal loss, whcrc 1/T2´
1/T2 + 2ry|oB(r)|. This lcaos to lowcr signaltonoisc in CPF imagcs comparco with SF imagcs. Sincc
TP is short comparco with T1, CPF imagcs tcno to havc T1 wcighting. Il TF approachcs T2´, thcn thc
T2´ wcighting bccomcs signilcant.
MR InstrumcntatIon
Thc kcy componcnts ol a moocrn MP imaging systcm incluoc a magnct, a pulsc scqucnccr, graoicnt ano
shim coils, ano an PI transmittcr/rcccivcr thc lunction ol which arc controllco by a host computcr
(Iigurc 79.8). To obtain thc B
0
lclo, most commcrcial scanncrs usc supcrconoucting magncts, although
somc spccialpurposc (ano oltcn lowcrcost) scanncrs may usc rcsistivc magncts. Supcrconoucting mag
ncts gcncrally havc lclo strcngths bctwccn 0.5 ano 1.0 T, whilc rcsistivc magncts normally havc lclo
strcngths < 0.3 T. Thc improvco signaltonoisc ratio obtainco with supcrconoucting ocsigns is ollsct by
thc ncco lor pcriooic cryogcn rcplaccmcnt. Morc moocrn magncts, howcvcr, minimizc this cost by
incluoing a cryogcn rcliquclcr. Two scts ol auxiliary graoicnt coils arc locatco within thc main magnct
M i M e M i M i
i i
z
T1
xy xy
T2
ano c
,
÷
, ,

,
÷ ÷
0
1 0
© 1999 by CRC Press LLC
to provioc spatially varying lclos, C(r,i), ano to allow shimming ol thc B
0
lclo. Currcnt graoicnt coil
harowarc can gcncratc maximum graoicnts ol up to 10 mT/m with risc timcs ol 120 µs ano allow lclos
ol vicw lrom 1 to 18 cm. Shim coils improvc thc homogcncity ol thc B
0
lclo by occrcasing oB(r) to a
lcw parts pcr million in orocr to minimizc T2´ cllccts ano spatial oistortions. Moocrn scanncrs incor
poratc a oigital PI subsystcm which cxcitcs thc spins ano thcn rccoros thc cmittco signals via onc or
morc PI coils within thc magnct. An PI synthcsizcr is couplco to both thc PI transmittcr ano rcccivcr,
so that synchronous octcction is possiblc. Thc PI systcm is conncctco cithcr to scparatc transmit ano
rcccivc coil(s) or to a combinco transmit/rcccivc coil(s).
To acquirc an MP imagc, thc host computcr intcracts with thc opcrator who oclncs thc imaging
paramctcrs (such as o, TP ano TF, slicc location, ano lclo ol vicw). Thc paramctcrs arc thcn translatco
into instructions which arc cxccutco on a synchronous statc machinc known as a pulsc scqucnccr. This
ocvicc proviocs rcaltimc control ol thc graoicnt ano PI wavclorms as wcll as othcr control lunctions,
such as unblanking thc PI rcccivcr ano cnabling thc ADC ouring an ccho. Data arc collcctco ano
ocmooulatco by thc rcccivcr, ano thcn imagcs arc rcconstructco using spccializco harowarc built normally
arouno a lastarray proccssor. Thc imagcs arc scnt to thc host computcr lor opcrator station oisplay,
archival or llming. In aooition, many MP scanncrs incorporatc ocviccs lor monitoring hcart ano
rcspiration ratc, ano allow thcsc signals to triggcr or gatc imagc acquisition. Iuturc MP imaging systcms
will probably incluoc highcr B
0
lclo ano graoicnts, ano lastcr oata proccssing/rcconstruction harowarc.
In aooition to currcnt imaging apparatus, it is likcly that ocoicatco instrumcnts will bc incrcasingly usco
to stuoy thc hcart ano lor pcrlorming ncurolunctional imaging ano MPguioco intcrvcntional proccourcs.
Rcfcrcnccs
1. C. Pcvcsz, H. I. Kunocl, ano M. A. Crabcr, Thc inlucncc ol structurco noisc on thc octcction ol
raoiologic abnormalitics, In\esi. IaJIoI., 9. 17918o, 1971.
2. A. Posc, VIsIon. Hunan anJ IIecironIc, Ncw York. Plcnum Prcss, 1973.
3. K. Doi, K. Strublcr, ano K. Possmann, Truncation crrors in calculating thc MTI ol raoiographic
scrccnllm systcms lrom thc linc sprcao lunction, I|ys. MeJ. ßIoI., 17. 211250, 1972.
1. Mooulation Translcr Iunction ol Scrccnllm Systcms, ICPU Pcport 11, 198o.
I!CURE 79.8 Ovcrvicw ol an MP scanncr showing kcy componcnts.
© 1999 by CRC Press LLC
5. H. Iujita, D.Y. Tsai, T. Itoh, K. Doi, ]. Morishita, K. Ucoa, ano A. Ohtsuka, A simplc mcthoo lor
octcrmining thc mooulation translcr lunction in oigital raoiography, IIII Trans. MeJ. Inag., 11.
3139, 1992.
o. ]. T. Dobbins III, D. I. Frgun, I. Putz, H. ßlumc, D. A. Hinshaw, ano D. C. Clark, DQF(l) ol lour
gcncrations ol computco raoiography acquisition ocviccs, MeJ. I|ys., 22. 15811593, 1995.
7. ]. M. ßoonc ano ]. A. Scibcrt, An analytical cogc sprcao lunction moocl lor computcr ltting ano
subscqucnt calculation ol thc ISI ano MTI, MeJ. I|ys., 21. 15111515, 1991.
8. ]. M. Sanorik ano P. I. Wagncr, Absolutc mcasurcs ol physical imagc quality. mcasurcmcnt ano
application to raoiographic magnilcation, MeJ. I|ys., 9. 510519, 1982.
9. P. C. ßunch, P. Shaw, ano P. I. van Mcttcr, Signaltonoisc mcasurcmcnts lor a scrccnllm systcm,
Iroc. SIII, 151. 1511o3, 1981.
10. P. C. ßunch, K. F. Hull, ano P. van Mcttcr, Analysis ol thc octcctivc quantum cllcicncy ol a
raoiographic scrccnllm combination, }. Cµi. Soc. An., 1. 902909, 1987.
11. ]. T. Dobbins III, Fllccts ol unocrsampling on thc propcr intcrprctation ol mooulation translcr
lunction, noisc powcr spcctra ano noisc cquivalcnt quanta ol oigital imaging systcms, MeJ. I|ys.,
22. 171181, 1995.
12. M. I. Cigcr, K. Doi ano C. F. Mctz, Invcstigation ol basic imaging propcrtics in oigital raoiography.
2. Noisc Wicncr spcctrum, MeJ. I|ys., 11. 797805, 1981.
13. P. ß. ßlackman ano ]. W. Tukcy, T|e Measureneni oI Iower Sµecira, Ncw York. Dovcr, 1958.
11. P. I. Wagncr ano ]. M. Sanorik, An introouction to oigital noisc analysis, in A. C. Haus, Fo., T|e
I|ysIcs oI MeJIcaI InagIng. IecorJIng Sysien Measurenenis anJ Tec|nIques, Ncw York. Amcrican
Association ol Physicists in Mcoicinc, pp. 521515, 1979.
15. C. T. ßarncs, Paoiographic mottlc. a comprchcnsivc thcory, MeJ. I|ys., 9. o5ooo7, 1982.
1o. P. I. Wagncr, Iast Iouricr oigital quantum mottlc analysis with application to rarc Farth intcnsi
lying scrccn systcms, MeJ. I|ys., 1. 1571o2, 1977.
17. H. H. ßarrctt ano W. Swinocll, IaJIoIogIcaI InagIng. Ncw York. Acaocmic Prcss, 1981.
18. C. M. ]cnkins ano D. C. Watts, SµeciraI AnaIysIs anJ Iis AµµIIcaiIons, San Irancisco, CA. Holocn
Day, 19o8.
19. K. Kocoooocr, ]. Strackcc, ano H. W. Vcncma, A ncw mcthoo lor microocnsitomctcr slit lcngth
corrcction ol raoiographic noisc powcr spcctra, MeJ. I|ys., 13. 1o9173, 198o.
20. M. I. Cigcr ano K. Doi, Invcstigation ol basic imaging propcrtics in oigital raoiography. 1. Moo
ulation translcr lunction, MeJ. I|ys., 11. 287295, 1981.
21. ]. T. ßushbcrg, ]. A. Scibcrt, F. M. Icioholot, ]r., ano ]. M. ßoonc, T|e IsseniIaI I|ysIcs oI MeJIcaI
InagIng, ßaltimorc, MD. Williams & Wilkins, 1991.
22. M. ßraun ano ß. C. Wilson, Comparativc cvaluation ol scvcral rarcFarth llmscrccn systcms,
IaJIoIogy, 111. 915919, 1982.
23. M. Sonooa, M. Takano, ]. Miyahara, ano H. Kato, Computco raoiography utilizing scanning lascr
stimulatco lumincsccncc, IaJIoIogy, 118. 833838, 1993.
21. W. Hillcn, U. Schicbcl, ano T. Zacngcl, Imaging pcrlormancc ol a oigital storagc phosphor systcm,
MeJ. I|ys., 11. 711751, 1987.
25. H. ßlumc, Stimulablc phosphor systcmstcchnical aspccts. In W. W. Pcpplcr ano A. Altcr, Fos.,
Iroc. C|esi InagIng ConI '87, Univcrsity ol Wisconsin, Maoison, pp. 191201, 1987.
2o. U. Ncitzcl, I. Maack, ano S. CunthcrKohlahl, Imagc quality ol a oigital chcst raoiography systcm
basco on a sclcnium octcctor, MeJ. I|ys., 21. 50951o, 1991.
27. C. H. Slump, C. ]. Iaanstra, H. Kuipcrs, M. A. ßocr, A. C. ]. Nijmcijcr, M. ]. ßcntum, P. Kcmncr,
H. ]. Mculcnbruggc, ano P. M. Snocrcn, Imagc quality charactcristic ol a novcl Xray octcctor with
multiplc scrccnCCD scnsors lor rcaltimc oiagnostic imaging. Iroc. SIII, 3032. o071, 1997.
28. I. F. Antonuk, Y. FlMohri, ]. H. Sicwcroscn, ]. Yorkston, W. Huang, V. F. Scarpinc, ano P. A.
Strcct, Fmpirical invcstigation ol thc signal pcrlormancc ol a highrcsolution, inoircct octcction,
activc matrix lat pancl imagcr (AMIPI) lor luoroscopic ano raoiographic opcration, MeJ. I|ys.,
21. 5170, 1997.
© 1999 by CRC Press LLC
29. D. I. Icc, I. K. Chcung, ano I. S. ]cromin, A ncw oigital octcctor lor projcction raoiography, Iroc.
SIII, 2132. 237219, 1995.
30. D. P ßoyo, P. C. Coulo, ]. P. Quinn, P. Sparks, ]. H. Stanlcy, ano W. ß. Hcrrmannslclot, A proposco
oynamic caroiac ocnsitomctcr lor carly octcction ano cvaluation ol hcart oiscasc, IIII Trans. NucI.
ScI., 2o. 27212727, 1979.
31. C. H. McCollough ano P. I. Morin, Thc tcchnical ocsign ano pcrlormancc ol ultralast computco
tomography, IaJIoI. CIIn. Nori| An., 32. 52153o, 1991.
32. P. M. Silvcrman, C. ]. Coopcr, D. I. Wcltman, ano P. K. Zcman, Hclical CT. practical consiocrations
ano potcntial pitlalls, IaJIograµ|Ics, 15(1). 253o, 1995.
33. ]. Hsich, A gcncral approach to thc rcconstruction ol Xray hclical computco tomography, MeJ.
I|ys., 23. 221229, 199o.
31. ]. Paoon, On thc octcrmination ol lunctions lrom thcir intcgrals along ccrtain manilolos, ßer.
Ver|. Sac|s. A|aJ. WIss. IeIµzIg, Mai|. I|ys. II., o9. 2o2277, 1917.
35. A. C. Kak ano M. Slancy, IrIncIµIes oI ConµuierIzeJ Tonograµ|Ic InagIng, Ncw York. IFFF Prcss, 1988.
3o. F. C. McCullough, ]. T. Paync, H. I. ßakcr, ]r., P. P. Hattcry, P. I. Shccoy, D. H. Stcphcns, ano F.
Ccogauous, Pcrlormancc cvaluation ano quality assurancc ol computco tomography scanncr, with
illustrations lrom thc FMI, ACTA, ano Dclta scanncrs, IaJIoIogy, 120. 173188, 197o.
37. F. C. McCullough ano ]. T. Paync, Xraytransmission computco tomography, MeJ. I|ys., 1. 8598,
1977.
38. P. T. Drocgc ano P. I. Morin, A practical mcthoo to mcasurc thc MTI ol CT scanncrs, MeJ. I|ys.,
9. 7587o0, 1982.
39. H. N. Wagncr, ]r., Z. Szabo, ano ]. W. ßuchanan, Fos., IrIncIµIes oI NucIear MeJIcIne, 2no co.,
Philaoclphia, PA. Saunocrs, 1995.
10. M. P. Sanolcr, Fo., DIagnosiIc NucIear MeJIcIne, 3ro co., ßaltimorc, MD. Williams & Wilkins, 199o.
11. P. Chanora, IniroJuciory I|ysIcs oI NucIear MeJIcIne, 1th co., Philaoclphia, PA. Ica & Icbigcr, 1992.
12. ]. A. Sorcnson ano M. F. Phclps, I|ysIcs In NucIear MeJIcIne, 2no co., Orlanoo, II. Crunc &
Stratton, 1987.
13. ]. I. Iittlclclo ano T. C. Puoo, {
99m
Tc]hyoroxymcthylcnc oiphosphonatc ano {
99m
Tc]mcthylcnc
oiphosphonatc biologic ano clinical comparison. concisc communication, }. NucI. MeJ., 21. o13, 1983.
11. C. ß. Saha, W. ]. MacIntyrc, P. C. ßrunkcn, P. T. Co, S. Paja, C. O. Wong, ano F. Q. Chcn, Prcscnt
asscssmcnt ol myocaroial viability by nuclcar imaging, SenIn. NucI. MeJ., 2o (1). 31535, 199o.
15. T. P. Millcr ano ]. W. Wallis, Caroiac nuclcar mcoicinc, Curr. Iro|I. DIagn. IaJIoI., 17 (5). 157193,
1988.
1o. P. ]. ßouorcau ano M. K. Iokcn, Iunctional imaging ol thc hcart, SenIn. NucI. MeJ., 17(1). 2838,
1987.
17. I. ]. Wackcrs, D. S. ßcrman, ]. Maooahi, D. D. Watson, C. A. ßcllcr, H. W. Strauss, C. A. ßouchcr,
M. Picaro, ß. I. Holman, ano P. Iriorich, Tcchnctium99m hcxakis 2mcthoxyisobutyl isonitrilc.
human biooistribution, oosimctry, salcty, ano prcliminary comparison to thallium201 lor myo
caroial pcrlusion imaging, }. NucI. MeJ., 30. 301311, 1989.
18. ß. M. W. Tsui, X. Zhao, F. C. Ircy, ano W. H. McCartncy, Quantitativc singlcphoton cmission
computco tomography. basics ano clinical consiocrations, SenIn. NucI. MeJ., 21. 38o5, 1991.
19. ]. Calowcll, D. Williams, C. Harp, ]. Stratton, ano ]. Pitchic, Quantitation ol sizc ol rclativc pcrlusion
oclcct by singlcphoton cmission computco tomography, CIrcuIaiIon, 70. 1018105o, 1981.
50. Pcrlormancc Mcasurcmcnts ol Scintillation Camcras, Stanoaros Publication No. NU11980, Wash
ington D.C.. National Flcctrical Manulacturcrs Association, 1980.
51. C. Muchllchncr, P. H. Wakc, ano P. Sano. Stanoaros lor pcrlormancc mcasurcmcnts in scintillation
camcras, }. NucI MeJ., 22. 7277, 1981.
52. Scintillation Camcra Acccptancc Tcsting ano Pcrlormancc Fvaluation, AAPM Pcport No. o, Ncw
York. Amcrican Association ol Physicists in Mcoicinc, 1980.
53. ComputcrAioco Scintillation Camcra Tcsting ano Pcrlormancc Fvaluation, AAPM Pcport No. 9,
Ncw York. Amcrican Association ol Physicists in Mcoicinc, 1987.
© 1999 by CRC Press LLC
51. Pcrlormancc Mcasurcmcnts ol Scintillation Camcras, Stanoaros Publication No. NU 1198o, Wash
ington D.C.. National Flcctrical Manulacturcrs Association, 198o.
55. Cuioc to Pcvisco Stanoaros lor Pcrlormancc Mcasurcmcnts ol Scintillation Camcras, Washington
D.C.. National Flcctrical Manulacturcrs Association, 198o.
5o. I. S. Craham, Quality assurancc ol Augcr camcras, in D. V. Pao, P. Chanora, ano M. C. Craham,
Fos., I|ysIcs oI NucIear MeJIcIne Ieceni AJ\ances, Monograph 10 ol thc Amcrican Association ol
Physicists in Mcoicinc (AAPM), Ncw York. Amcrican Institutc ol Physics, pp. o882, 1981.
57. I. A. O`Tuama ano S. T. Trcvcs, ßrain singlcphoton cmission computco tomography lor bchavior
oisorocrs in chilorcn, SenIn. NucI. MeJ., 23 (3). 2552o1, 1993.
58. C. ß. Saha, W. ]. MacIntyrc, ano P. T. Co, Paoiopharmaccuticals lor brain imaging, SenIn. NucI.
MeJ., 21 (1). 321319, 1991.
59. S. Stcicn ano ]. Aascth, Thallium201 as an agcnt lor myocaroial imaging stuoics, AnaIysi, 120 (3).
779781, 1995.
o0. A. Taylor, ]r. ano ]. V. Nally, Clinical applications ol rcnal scintigraphy, An. }. IoenigenoI., 1o1.
3111, 1995.
o1. P. Schlosscr ano S. Schlcgcl, D2rcccptor imaging with {123I]IßZM ano singlc photon cmission
tomography in psychiatry. a survcy ol currcnt status, }. NeuraI Transn. Gen. Seci., 99 (13).
173185, 1995.
o2. T. Sunocrlano, C. Fsposito, S. F. Molchan, P. Coppola, D. W. ]oncs, ]. Corcy, ]. T. Iittlc, M. ßahro,
ano D. P. Wcinbcrgcr, Dillcrcntial cholincrgic rcgulation in Alzhcimcr`s paticnts comparco to
controls lollowing chronic blockaoc with scopolaminc. a SPFCT stuoy, Isyc|oµ|arnacoIogy, 121
(2). 231211, 1995.
o3. ß. Shapiro, ]. C. Sisson, ß. I. Shulkin, M. D. Cross, ano S. Zcmpcl, Thc currcnt status ol mcta
iooobcnzylguanioinc ano rclatco agcnts lor thc oiagnosis ol ncurocnoocrinc tumors, Ç. }. NucI.
MeJ., 39 (1 Suppl. 1). 38, 1995.
o1. ß. Shapiro, Imaging ol catccholamincsccrcting tumours. uscs ol MIßC in oiagnosis ano trcatmcnt,
ßaIIIIeres CIIn. InJocrInoI. Meia|oI., 7 (2). 191507, 1993.
o5. M. P. Stokkcl, ß. M. Kwa, ano F. K. Pauwcls, Imaging ano staging ol smallccll lung canccr. is thcrc
a luturc rolc lor octrcotioc scintigraphy¹ ßr. }. CIIn. Iraci., 19. 235238, 1995.
oo. F. P. Krcnning, D. ]. Kwckkcboom, ]. C. Pcubi, P. M. Van Hagcn, C. H. van Fijck, H. Y. Oci, ano
S. W. Iambcrts, {
111
In]octrcotioc scintigraphy in oncology, Meia|. CIIn. Ixµ., 11 (9 Suppl. 2).
838o, 1992.
o7. F. ]. Hollman ano M. F. Phclps, Positron cmission tomography. principlcs ano quantitation, in
M. Phclps, ]. Mazziotta, ano H. Schclbcrt, Fos., IosIiron InIssIon Tonograµ|y anJ AuioraJIograµ|y.
IrIncIµIes anJ AµµIIcaiIons Ior i|e ßraIn anJ Heari, Ncw York. Pavcn Prcss, pp. 23728o, 198o.
o8. P. A. Kocppc ano C. D. Hutchins, Instrumcntation lor positron cmission tomography. tomograph
ano oata proccssing ano oisplay systcms, SenIn. NucI. MeJ., 22 (3). 1o2181, 1992.
o9. M. F. Cascy ano P. Nutt, A multicrystal two oimcnsional ßCO octcctor systcm lor positron
cmission tomography, IIII Trans. NucI. ScI., NS33. 1o01o3, 198o.
70. C. I. Mclchcr ano ]. S. Schwcitzcr, Ccriumoopco lutctium oxyorthosilicatc. a last, cllcicnt ncw
scintillator, in A|siracis oI Iaµers IresenieJ ai IIII NucIear ScIence SynµosIun, Santa Ic, NM,
Novcmbcr, 1991 (Abstract).
71. K. Wicnharo, M. Dahlbom, I. Friksson, C. Michcl, T. ßruckbaucr, U. Pictrzyk, ano W. D. Hciss,
Thc FCAT FXACT HP. pcrlormancc ol a ncw high rcsolution positron scanncr, }. Conµui. AssIsi.
Tonogr., 18. 110118, 1991.
72. T. P. DcCraoo, T. C. Turkington, ]. ]. Williams, C. W. Stcarns, ]. M. Hollman, ano P. F. Colcman,
Pcrlormancc charactcristics ol a wholc booy PFT scanncr, }. NucI. MeJ., 35. 1398110o, 1991.
73. ]. A. Karp, C. Muchllchncr, D. A. Mankoll, C. F. Orooncz, ]. M. Ollingcr, M. F. DaubcWithcrspoon,
A. T. Haigh, ano D. ]. ßccrbohm, Continuousslicc PFNNPFT. a positron tomograph with volumc
imaging capability, }. NucI. MeJ., 31. o17o27, 1990.
© 1999 by CRC Press LLC
71. C. ]. Pavlin, K. Harasicwicz, M. D. Shcrar, ano I. S. Iostcr, Clinical usc ol ultrasouno biomicroscopy,
Cµ|i|aInoIogy, 98. 287295, 1991.
75. D. H. Turnbull, High lrcqucncy ultrasouno imaging, in I. Coloman ano ]. ß. Iowlkcs, Fos., MeJIcaI
CT anJ UIirasounJ, Curreni Tec|noIogy anJ AµµIIcaiIons, Maoison, WI. Aovancco Mcoical Pub
lishing, pp. 285297, 1995.
7o. P. N. T. Wclls, Propagation ol ultrasouno wavcs through tissuc, in C. D. Iullcrton ano ]. A. Zagzcbski,
Fos., MeJIcaI I|ysIcs oI CT anJ UIirasounJ. TIssue InagIng anJ C|aracierIzaiIon, AAPM Monograph
o, Collcgc Park, MD Amcrican Association ol Physicists in Mcoicinc, pp. 3o7387, 1980.
77. ]. A. Zagzcbski, IsseniIaIs oI UIirasonIc I|ysIcs, St. Iouis, MO. Mosby, 199o.
78. K. Shung, In \Iiro cxpcrimcntal rcsults on ultrasonic scattcring in biological tissucs, in K. Shung
ano C. Thicmc, Fos., UIirasonIc ScaiierIng In ßIoIogIcaI TIssues, ßoca Paton, II. CPC Prcss,
pp. 291312, 1993.
79. M. Insana, Souno attcnuation in tissuc, in I. Coloman ano ]. ß. Iowlkcs, Fos., MeJIcaI CT anJ
UIirasounJ, Curreni Tec|noIogy anJ AµµIIcaiIons, Maoison, WI. Aovancco Mcoical Publishing,
pp. 1533, 1995.
80. D. H. Turnbull, Iunoamcntals ol Acoustic Transouction, in I. Coloman ano ]. ß. Iowlkcs, Fos.,
MeJIcaI CT anJ UIirasounJ, Curreni Tec|noIogy anJ AµµIIcaiIons, Maoison, WI. Aovancco Mcoical
Publishing, pp. 19oo, 1995.
81. K. Icrrara ano C. DcAngclis, Color low mapping, UIirasounJ MeJ. ßIoI., 23(3). 321315, 1997.
82. K. ]. W. Taylor, P. N. ßurns, ano P. N. T. Wclls, CIInIcaI AµµIIcaiIon oI DoµµIer UIirasounJ, 2no co.,
Ncw York. Pavcn Prcss, pp. 1931, 5591, 1995.
83. SianJarJ Mei|oJs Ior MeasurIng i|e IerIornance oI IuIse Ic|o UIirasounJ InagIng IquIµneni,
Iaurcl, MD. Amcrican Institutc ol Ultrasouno in Mcoicinc, 1991.
81. H. Iopcz, M. H. Iocw, P. I. ßutlcr, M. C. Hill, ano P. M. Allman, A clinical cvaluation ol contrast
octail analysis lor ultrasouno imagcs, MeJ. I|ys., 17. 1857, 1993.
85. Mei|oJs Ior MeasurIng IerIornance oI IuIseIc|o UIirasounJ IquIµneni Iari II. DIgIiaI Mei|oJs,
Iaurcl, MD. Amcrican Institutc ol Ultrasouno in Mcoicinc, 1995.
8o. ]. ]. Powno, F. I. Maoscn, ]. A. Zagzcbski, C. P. Irank, ano I. Dong, Phantoms ano automatco systcm
lor tcsting thc rcsolution ol ultrasouno scanncrs, UIirasounJ MeJ. ßIoI., 23(2). 2152o0, 1997.
87. SianJarJ SµecIfcaiIon oI Ic|oscoµe SensIiI\Iiy anJ NoIse Ie\eI IncIuJIng IeconnenJeJ IraciIce Ior
Suc| Measurenenis, Iaurcl, MD. Amcrican Institutc ol Ultrasouno in Mcoicinc, 1991.
88. ÇuaIIiy Assurance ManuaI Ior Gray ScaIe UIirasounJ Scanners, Iaurcl, MD. Amcrican Institutc ol
Ultrasouno in Mcoicinc, 1995.
89. C. M. Pumack, S. P. Wilson, ano ]. W. Charboncau, DIagnosiIc UIirasounJ, 2no co., St. Iouis, MO.
Mosby, 1997.
90. P. I. Carson ano M. M. Cooositt, Pulsco ccho acccptancc ano quality control tcsting. In I. Coloman
ano ]. ß. Iowlkcs, Fos., MeJIcaI CT anJ UIirasounJ, Curreni Tec|noIogy anJ AµµIIcaiIons, Maoison,
WI. Aovancco Mcoical Publishing, pp. 15519o, 1995.
91. C. P. Slichtcr, IrIncIµIes oI MagneiIc Iesonance, ßcrlin. SpringcrVcrlag, 1980.
92. I. ßloch, Nuclcar inouction, I|ys. Ie\., 70. 1o0185, 191o.
93. W. A. Foclstcin, ]. M. S. Hutchinson, C. ]ohnson, ano T. Pcopath, Spinwarp imaging ano appli
cations in wholc booy imaging, I|ys. MeJ. ßIoI., 25. 75175o, 1980.
91. W. S. Hinshaw ano A. H. Icnt, An introouction to NMP imaging. Irom ßloch cquation to thc
imaging cquation, Iroc. IIII, 71. 338350, 1983.
Furthcr InformatIon
P. N. ßraccwcll, T|e IourIer TransIorn anJ Iis AµµIIcaiIons, 2no co., Ncw York. McCrawHill, 1978.
F. O. ßrigham, T|e Iasi IourIer TransIorn, Fnglcwooo Clills, N]. PrcnticcHall, 1971.
M. A. ßrown ano P. C. Scmclka, MII. ßasIc IrIncIµIes anJ AµµIIcaiIons, Ncw York. WilcyIiss, 1995.
© 1999 by CRC Press LLC
]. C. Dainty ano P. Shaw, Inage ScIence, Ncw York. Acaocmic Prcss, 1971.
F. Iukushima ano S. ß. W. Pococr, IxµerIneniaI IuIse NMI. A Nuis anJ ßoIis Aµµroac|, Pcaoing, MA.
AooisonWcslcy, 1981.
H. K. Huang, IIenenis oI DIgIiaI IaJIoIogy, Fnglcwooo Clills, N]. PrcnticcHall, 1987.
A. Macovski, MeJIcaI InagIng Sysiens, Fnglcwooo Clills, N]. PrcnticcHall, 1983.
P. Sprawls ano M. ]. ßronskill, Fos., T|e I|ysIcs oI MII. 1992 AAIM Sunner Sc|ooI IroceeJIngs,
Wooobury, NY. Amcrican Association ol Physicists in Mcoicinc, 1993. {An cxccllcnt ovcrvicw ol
currcnt MP instrumcntation ano clinical applications writtcn by cxpcrts in thc lclo.]

Sponsor Documents

Recommended

No recommend documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close