Evaluation of a Thyroid Fluorescent Scanning System

of Concentric Source—Detector Design

Michael T. GiIlin,* James H. Thrall,t Robert J. Corcoran, and Merrill C. Johnson

Walter Reed Army Medical Center, Washington, D.C.

A concentric source—detector system for thyroid fluorescent scanning isdescribed, including fundamental parameters of system response and adaptation of a conventional rectilinear scanner for use with it. The basic systemconsists of twenty 1-Ci sources of 241Am, a 500-mm2 Si(Li) detector, and

associated pulse-height electronics. The image-forming equipment of therectilinear scanner is retained. We have developed a clinical imaging tech

nique that provides a photon density of 600—800 counts/cm2 over the thyroid gland in subjects with normal iodine pools. Comparisons are made

between the outrigger design for fluorescent scanning and conventionalemission scanning.

J Nuci Med 18: 163—167, 1977

Over the past 6 years two geometrically differentdesigns for thyroid fluorescent scanning systems havebeen developed (1—6). In one, a single source of241Am is positioned in an outrigger configurationwith respect to the detector (2—4). In the other design, several sources of 241Amare arranged concentrically around a centrally placed detector (5,6).We have converted a conventional rectilinear scanner for use with the first commercial version of theconcentric design. The basic parameters of this sys

tem's response have been determined and appliedto the development of clinical imaging techniques.

MATERIALS AND METHODS

The fluorescent scanning head consists of a leadcollimator and source holder* containing 20 individual 1-Ci sourcest of 241Am, and a 500-mm2 Si(Li)detectors with a 10-liter gravity-feed liquid-nitrogen(LN2) Dewar. The 241Am sources are located in theholder to form a concentric circle around the detector. Each source is collimated through a taperedhole aimed at a common point 4.5 cm from the faceof the collimator. The enclosed detector is collimatedby a single hole tapered to the same point. The axesof the exciting beams form a 30° angle with the

central axis of the detector.The fluorescent scanning head was mounted on

the probe arm of a scanner@ from which the standard detector and collimator were removed. The

weight of the head of the fluorescent scanning unit(lead collimator, source holder, detector, and 10-liter Dewar) is 150 pounds, which is less than theI 85-pound combined weight of the Nal detector andcollimator that the scanner is designed to support.A heavy aluminum bracket attaches to the base ofthe vertical arm of the scanner and extends in horseshoe-like fashion over the top of the collimator andsource holder (Fig. I ) . Several bolts pass throughthe bracket, the top plate, and the main body ofthe collimator and source holder. The preamplifiersits in the opening of the horseshoe bracket. Thissystem retains all motions of the scanner arm excepttilting.

The 500-mm2 Si(Li ) detector has a sensitive depthof approximately 5.4 mm, which yields an intrinsicefficiency of 80—100% for the 28.5-keV K-alphacharacteristic x-ray of elemental iodine. It is possibleto temperature cycle the Si(Li) detector, provided

the biasing voltage has been turned off. The pulse

Received May 29, 1975; revision accepted Sept. 22, 1976.For reprints contact: James H. Thrall, Nuclear Medicine

Sect., University of Michigan Medical Center, 1405 E. AnnSt., Ann Arbor, MI 48109.

* Present address: Radiation Therapy Dept., Medical Col

lege of Wisconsin, Milwaukee, WI 53226..tPresentaddress:NuclearMedicineSect.,,Dept. of In

ternal Medicine, University of Michigan Medical Center,Ann Arbor, MI 48109.

Volume 18, Number 2 163

by on January 15, 2016. For personal use only. jnm.snmjournals.org Downloaded from

GILLIN, THRALL, CORCORAN, AND JOHNSON

FIG. 1. Photographof system.Irradi.ator—detectordevice has replaced conven@tional scanner head.

height analyzer, which must be capable of providinggain stability and high resolution at high count rates,consists of a preamplifier with LN2-cooled field-effecttransistor, an amplifier with pole-zero cancellation

and baseline restoration,@@ and a dc-coupled singlechannel analyzer.@

The output pulse of the single-channel analyzer(SCA) is a positive 5-V signal; it is attenuated by afactor of 10 and then inverted before being fed into

the pulse-control section of the scanner. No modifications of the scanner's electronics were needed.The normal scanner controls are used to form aphotoscan and a paper dot scan. A technique for

clinical imaging was developed to yield a maximumphoton density of 600—800 counts/cm2 while scanning over a normal thyroid gland. A series of patientswith a variety of thyroid abnormalities undergoingconventional emission studies have also been imagedwith scan factors held constant to allow qualitativecomparison of iodine pools (4,6).

The pulse-height spectrum representing 0—60keVwas studied with a multichannel analyzer to determine the appropriate SCA window setting and to

assess the effects of scattered radiation on the signalto-noise ratio for the chosen window. The full widthat half maximum (FWHM) and the full width attenth maximum (FWTM) of the K-alpha peak weremeasured in tissue-equivalent neck phantoms and

in clinical studies.The width of the exciting radiation beam was

measured with film both in air and in the tissue

equivalent phantom. The overall system responsewas determined by measuring line spread functions

using a I -mm-diam line-source of elemental iodine.This was first placed on the surface of a tissueequivalent phantom simulating the neck. Line spread

functions were obtained at collimator-to-line-source

distances of 3.5, 4.5, and 5.5 cm. The line spreaddeterminations were then repeated at collimator-toline-source distances of 2.5, 3.5, 4.5, and 5.5 cm,with the tissue-equivalent phantom behind the sourceand an additional 1.5 cm of tissue-equivalent material placed immediately over the source.

The exposure rate at the focal point was determined using film and thermoluminescent dosimeters.The radiation absorbed dose for the thyroid glandper imaging procedure was estimated for the expo

sure rate and clinical scanning technique, and wasmeasured in phantoms by thermoluminescent dosimetry.

RESULTS

The pulse-height spectrum obtained when a sourceof stable iodine is scanned in air contains three prominent peaks: the iodine K-alpha x-ray peak at 28.5

keV, the K-beta peaks at around 32.5 keV, and aI 50°Compton scatter peak at 48.5 keV, due to the59.5-keV gamma of 241Am.When the iodine sourceis contained in a phantom, scattered radiation increases significantly and the K-beta peak becomeslost in the Compton-scatter background. Over the

thyroid gland in normal subjects, the K-alpha peakto-background ratio approaches 5. The FWHM ofthe K-alpha peak, as measured in the pulse-heightspectrum obtained during scans on normal subjects,is approximately I keV. The FWTM is I .8 keV, andthis was chosen as the window width for the SCAand was centered on the iodine K-alpha x-ray peakfor imaging purposes. The count rate in this windowwhen the scanner is over the gland in normal subjects is approximately 60—120 cps, while the countrate in the entire spectrum ranges from 5,000 to

164 JOURNAL OF NUCLEAR MEDICINE

by on January 15, 2016. For personal use only. jnm.snmjournals.org Downloaded from

INSTRUMENTATION AND PHYSICS

gland is a direct function of imaging time, and forthe scanning technique described below it measuredless than 50 mrads per scan. Note that only the thyroid gland and soft tissues surrounding it receive thisdose with the fluorescence technique ( 1.—6).

For clinical studies, a line spacing of 2 mm isused. The linear scanning speed is set at 50 cm/mmto meet the objective of a photon density of 600—800 counts/cm2 over normal glands. In normal subjects the usual scan width is 8—9cm and the heightis 6—8cm. Patients are positioned as for conventionalemission scanning, and study times average 7—9mmafter setup. The representative studies illustrated in

Fig. 4 were obtained as follows: in an athyreoticpatient with no detectable iodine (Fig. 4A) ; and inpatients with total thyroidal iodine judged to bedecreased (Fig. 4B), normal (Fig. 4C), and increased (Fig. 4D).

DISCUSSION

In fluorescent scanning, as in emission scanning,DIstanceFrom the choice of window width for the SCA that is set

@:0o:1:,.F:of for the photopeak of interest is a compromise be

tween sensitivity and resolution. A window widthequal to the FWTM for the iodine K-alpha x-raypeak was felt to provide the best compromise withthis fluorescent system. The minimally improvedresolution observed experimentally with narrowerwindows was offset by decreased sensitivity andprolonged imaging times for clinical studies.

The iodine K-alpha x-ray-to-background ratio is

24

22

20

Th@ @.@

FOCALPLANE:FIHN 0.1caFITN I.5c.

RELATIVE

OFCOuNTS

RELATIVENUNNER

OFCOUNTS

cmI 23

(cm)

FIG.2. Linespreadfunctionsobtainedwithiodinelinesourceon surface of 15-cm-thick tissue.equivalent neck phantom. Distancesare from collimator face to line source.

20,000 cps. The inherent background activity (roomand system) at the energy of the iodine K-alpha x-rayis less than 5 cps.

The FWTM of the exciting radiation beam at thefocal point in air is 1.2 cm. With the film at a 1.5-cm depth below the surface of the phantom, thisFWTM increases to 1.4 cm at the focal point.

Line spread functions with the phantom behindthe source are illustrated in Fig. 2. The doublepeaking at 3.5 cm indicates that the exciting radiationbeams have not reached the focal point at whichthey converge. The line spread functions in Fig. 3simulate the clinical situation, with scattering material both in front of and behind the line source.Divergence of the individual exciting beams beyondthe focal point is indicated by the double peakingat 5.5 cm. The FWHM of the line spread functionsin the focal plane is approximately equal for bothscattering situations: 0.7 cm without overlying material and 0.8 cm with 1.5 cm of overlying scatteringmaterial. The corresponding FWTM, however, ismuch greater in the latter situation because of thescatter (Figs. 2 and 3).

The exposure rate at the focal plane is 50 mrads/mm. The radiation absorbed dose to the thyroid

20

‘a13 cmPhantom@ FrontofSourcea@dThictPhonlom

FOCAL.PLANE:FIHN-O.8cmFITN'lO.O cm

i6

14

12

10

8

6

4

2

D@onceftomFrootFoceofCOIIMIOtOr

(cm)

FIG.3. Linespreadfunctionsobtainedwithiodinelinesource,1.5 cm below surface of 15-cm-thick neck phantom. Distances arefrom collimator face to line source.

Volume 18, Number 2 165

by on January 15, 2016. For personal use only. jnm.snmjournals.org Downloaded from

GILLIN, THRALL, CORCORAN, AND JOHNSON

inherent tomographic properties, the line spreadfunctions in Fig. 3 indicate a response within 70%of the maximum over a depth of 2 cm. This is similarto the response observed by Patton et al. (6) . TheFWHM of the line spread function in the focal plane(in the simulated clinical situation) compares favorably with similar determinations from conventionalrectilinear scanners, but ultimate system resolutionis not as good as may be achieved with a scintillationcamera and a pinhole collimator.

The radiation absorbed dose of 50 mrads perfluorescent scan is comparable with that receivedby the thyroid from an emission scan with 5 mCi of

: pertechnetate,andisfarlessthantheradiationabsorbed dose from 30 @@Ciof 131! or 1 mCi of 1@I(8,9). With the fluorescent system, the radiation

absorbed dose is limited to the neck, and the wholebody dose is essentially zero.

It is useful to maintain the scan factors constant: influorescentimaging(4—6). Thisallowsdirect

qualitative interscan comparison of thyroidal iodine: pools. This is clearly illustrated in Fig. 4, although

it is important not to confuse content with concentration. The glands in Figs. 4C and 4D may wellhave the same concentration but the total iodine poolin the latter is larger because of the larger glandsize. In addition to the qualitative use of fluorescentscanning, Patton et al. (6) have shown a quantita

tive capability for these systems to determine thyroidal iodine content in milligrams.

The scanning time of 7—9mm for the system issubstantially faster than that quoted for the outrigger design (4) . The greater quantity of 241Amthat is used in the concentric-source design is partlyresponsible and there is probably less self-absorbtionin the 20 individual sources than in the single largesource used in the outrigger design.

CONCLUSIONS

Fluorescent scanning provides the only directmeans of studying thyroidal iodine content and distribution in vivo. Further design improvements willundoubtedly be forthcoming, but our experiencesuggests that with appropriate calibration, a currently available concentric-source fluorescent systemprovides good images of the thyroidal iodine pool inreasonable times and with low radiation dose to thepatient.

FOOTNOTES

* Ortec-482 (Oak Ridge, Tenn.).

t Monsanto Model 2704 thin-window gamma sources.@ Ortec-7616-25-350.

§PickerMagnascannerV (Cleveland,Ohio).IIOrtec-433A.¶

A .@. i... B. ..

-@@ : .@@ .

•0@ :@.mii::'! @:@

C 0

FIG.4. Typicalfluorescentscans:Athyreoticpatientwithnodetectable iodine (A), and patients with low (B), normal (C), andincreased (D) total thyroidal iodine pools.

a significant determinant of scan quality in fluorescent imaging. The observed ratios of up to 5, whenthe exciting radiation beam is over normal thyroidglands, are not unfavorable when compared to emission studies using 99@'Tcas pertechnetate. For thisthe ratio of gland-to-background counts is often lessthan 5. It should be noted, however, that the natureof the background is different in the two types ofscanning. In fluorescent scanning, the backgroundconsists of Compton-scattered photons from the exciting radiation beam. In emission scanning theanalogous background consists of primary and scattered photons from extraglandular tracer and scattered photons originating within the gland.

An important signal-processing problem in fluorescent imaging is the combination of a high integralcount rate in the entire spectrum and a low photopeak fraction. During an entire scan, far less than1% of the total input to the SCA is accepted. Thecomparable figure ranges from 20 to 33 % for emission scanning using the scintillation camera and

9OmTc_labeled tracers (7) . The pulse-height electronics used in a fluorescent system must be capableof handling high count rates. Initially we used anac-coupled SCA with this system. It suffered froma window shift between high and low input fluxesand was replaced by the dc-coupled SCA.

Although fluorescent scanning systems have some

166 JOURNAL OF NUCLEAR MEDICINE

by on January 15, 2016. For personal use only. jnm.snmjournals.org Downloaded from

INSTRUMENTATION AND PHYSICS

Vanderbilt. In Semiconductor Detectors in Medicine. OakRidge, Tenn., USAEC Conf-730321, 1973, pp 254—294

6. PATrON JA, HOLLIFIELD JW, BRILL AB, et al. : Differentiation between malignant and benign solitary thyroidnodules by fluorescent scanning. I Nucl Med 17: 17—21,1976

7. ARNOLDJE, JOHNSTONAS, PINSKY SM : Resolvingtime of scintillation cameras. I NucI Med 15: 475, 1974

8. ATKINS HL: Technetium-99m pertechnetate uptakeand scanning in the evaluation of thyroid function. SeminNuciMed 1:345—355,1971

9. WELLMAN HN, ANGER RT: Radionuclide dosimetryand the use of radioiodines other than @@‘Iin thyroid diagnosis. Semin Nuci Med I : 356—378,1971

REFERENCES

1. HOFFERPB, JONESWB, CRAWFORDRB, et al.: Fluorescent thyroid scanning: A new method of imaging thethyroid. Radiology 90: 342—344,1968

2. HOFFERPB: Fluorescent thyroid scanning. Am IRoentgenol105: 721—727,1969

3. HOFFER PB, GOTTSCHALKA: Fluorescent thyroid scanfling: Scanning without radioisotopes. Radiology 99: 117—123, 1971

4. HOFFER PB, BERNSTEIN J, GOTrSCHALK A : Fluorescenttechniques in thyroid imaging. Semin Nucl Med 1: 379—389,1971

5. PATTON JA, BRILL AB, BLANCO J, et al. : Experienceswith semiconductors in imaging and function studies at

October 21—23,1977 Aladdin Hotel Las Vegas, Nevada

ANNOUNCEMENTAND CALLFOR ABSTRACTSFOR SCIENTIFICPROGRAM

TheScientificProgramCommitteewelcomesthe submissionof abstractsof originalcontributionsin nuclear medicine from members and nonmembers of the Society of Nuclear Medicine for the 2nd AnnualWestern Regional Meeting. Physicians, Scientists, and Technologists, members and nonmembers are invitedto participate. The Program will be structured to permit the presentation of papers from all areas of interestin the specialty of Nuclear Medicine. A separate set of guidelines will be used to judge the Technologist abstracts. Abstracts for the scientific program will be printed in the program booklet and will be available toall registrants at the meeting.

GUIDELINESFOR SUBMITTINGABSTRACTS

The abstracts will be printed from camera-ready copy provided by the authors. Therefore, only abstractsprepared on the official abstract form will be considered. These abstract forms will be available from theWestern Regional Chapter's SNM office (listed below). Abstract forms will only be sent to SNM members ofthe Pacific Northwest, Southern California, Northern California, and Hawaii Chapters in a regular mailing. Allother requestswill be sent on an individual basis.

All participants will be required to register and pay the appropriate fee.

Please send the original abstract form, supporting data, and six copies to:

Jean Lynch, Administrative Coordinator2nd Western Regional MeetingP.O. Box 40279San Francisco, CA 94140

Deadline for abstract submission: Postmark midnight, July 8, 1977.

Volume 18, Number 2 167

2nd ANNUAL WESTERN REGIONAL MEETING

THE SOCIETY OF NUCLEAR MEDICINE

by on January 15, 2016. For personal use only. jnm.snmjournals.org Downloaded from

1977;18:163-167.J Nucl Med.   Michael T. Gillin, James H. Thrall, Robert J. Corcoran and Merrill C. Johnson  DesignEvaluation of a Thyroid Fluorescent Scanning System of Concentric Source-Detector

http://jnm.snmjournals.org/content/18/2/163This article and updated information are available at:

  http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:  

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 1977 SNMMI; all rights reserved.

by on January 15, 2016. For personal use only. jnm.snmjournals.org Downloaded from