Radiation dose reduction in diagnostic neonatal x-ray imaging

Groenewald, Annemari (2013-03)

Thesis (MScMedSc)--Stellenbosch University, 2013.

Thesis

ENGLISH ABSTRACT: Introduction - X-ray imaging is used to diagnose and follow up various conditions in neonates (i.e. pre-term babies and babies up to the age of 28 days). Chest anterior-posterior (AP) radiographs are used to check the condition of the lungs and heart. Acceptable images requiring lower doses of radiation can be produced digitally by using a computed radiography (CR) system. Radiation can induce cancer in the young child. Lower doses are therefore important since neonates are more sensitive to radiation and have a relatively longer life expectancy. To minimise the risk of inducing cancer in neonates, x-ray exposures must adhere to the principle of ALARA (i.e. as low as reasonably achievable). Reducing radiation doses during a radiographic examination of a neonatal chest often results in reduced image quality. Dose reduction while maintaining optimum image quality and the risk of inducing cancer must therefore be considered in conjunction with one another. Aim - The aim of this study is to develop an anatomical and radiological simulation phantom of a real neonatal chest and, using the phantom, to derive methods of decreasing the radiation dose while maintaining acceptable quality of the clinical image at a reduced cancer induction risk. Materials and methods - Following guidelines in literature on the subject, as well as principles of medical physics, a phantom simulating a real neonatal chest anatomically and radiologically was developed. Anatomical equivalence was based on a computed tomography (CT) scan of a neonatal cadaver. Radiological equivalence was obtained by matching density, elemental composition, attenuation, scatter and absorption characteristics of real neonatal tissues to possible substitute materials. The phantom was used to derive x-ray imaging protocols to decreased radiation dose, as well as the risk of cancer induction, while maintaining acceptable quality of the image. To achieve this exposure technique factors, such as tube voltage and current, exposure time and filtration, were varied experimentally. Image quality was evaluated quantitatively in a physics image quality assessment phantom, by calculating signal-to-noise ratios and modulation transfer functions. Images were ranked according to measured doses, visual and quantitative image quality and cancer induction risks. Results - The simulation phantom acceptably matched a real neonatal chest anatomical and radiologically. The radiation dose and image quality of various exposures were compared with the standard exposure for neonatal chest AP x-ray imaging. In eight different exposures the dose was decreased to below the standard. The largest dose reduction was approximately 63%. Seven of these images had an improved visual image quality compared with the standard. The greatest improvement being about 21%. In two of the eight options a cancer induction risk analysis showed that, despite reduced doses, the risk could be greater than the standard exposure risk. Discussion and conclusion - In six different exposure options a decrease in the dose was achieved while maintaining, and even improving, image quality and lowering the cancer inducing risk. These exposure protocols were recommended to be used in the Diagnostic Radiology Division of Tygerberg Academic Hospital.

AFRIKAANSE OPSOMMING: Inleiding- X-straalbeelding word gebruik om verskeie siektes te diagnoseer en op te volg. In die geval van neonate (i.e. vroeg-gebore babas en babas van geboorte tot ouderdom 28 dae) word anterior-posterior (AP) x-straalondersoeke van die borskas gedoen om die toestand van die longe en hart te evalueer. Bevredigende beeldkwaliteit wat laer dosisse x-strale nodig het, kan digitaal verkry word met behulp van ʼn rekenaar-radiografiese (i.e. Computed Radiography (CR)) sisteem. By die jong kind kan blootstelling aan bestraling kanker veroorsaak. In die geval van neonatale blootstellings is dit belangrik, want neonate is meer sensitief vir bestraling en het relatief ʼn langer lewensverwagting. Om die risiko vir kankerinduksie te minimaliseer moet daar aan die beginsel van ALARA (“as low as reasonably achievable”) voldoen word. Dosisverlagings gaan gewoonlik gepaard met verlies van beeldkwaliteit. Die doelwit tydens die radiografiese ondersoek van ʼn neonaat moet egter altyd verlaging van die bestralingsdosis met optimum beeldkwaliteit en verlaagde risiko van kankerinduksie in die neonaat wees. Doelwit - Die doelwit van hierdie studie is om ʼn gesimuleerde fantoom te ontwikkel wat ʼn regte neonatale borskas anatomies en radiologies voorstel. Deur die fantoom te gebruik word metodes ontwikkel om bestralingsdosisse te verminder met die doel om die kankerinduksierisiko te verlaag, maar steeds aanvaarbare visuele beeldkwaliteit te behou. Materiaal en metodes - Riglyne en aanbevelings uit literatuur oor die onderwerp en basiese mediese fisika beginsels is aangewend om ʼn fantoom te maak wat ʼn neonatale borskas anatomies en radiologies naboots. Anatomiese ekwivalensie is verkry deur ʼn neonatale kadawer rekenaar-tomografies te skandeer. Radiologiese ekwivalensie is behaal deur digtheid, elementale komposisie, attenuasie, absorpsie en verstrooiing-karakteristieke van ‘n regte neonatale borskas te vergelyk met dié van moontlike vervangingsmateriale. Die fantoom is gebruik om x-straalbeeldprotokolle af te lei wat die bestralingsdosis en die kankerinduksierisiko verlaag terwyl die beeldkwaliteit behoue bly. Dit is bereik deur blootstellingsfaktore, soos x-straalbuisspanning en stroom, blootstellingstyd en filtrasie, eksperimenteel te varieer. Beeldkwaliteit is ook kwantitatief ge-evalueer in ʼn fisikafantoom, deur sein-tot-geruisverhoudings en modulasie-oordragsfunksies te bereken. Die beelde is georden volgens gemete ingangsdosisse, visuele en kwantatiewe beeldkwaliteit en kankerinduksierisiko. Resultaat - Die simulasiefantoom van die neonatale borskas was ʼn aanvaarbare anatomiese en radiologiese voorstelling van ‘n ware neonatale borskas. Die bestralingsdosis en beeldkwaliteit van verskillende blootstellings is vergelyk met die standaard blootstelling wat gewoonlik op neonate tydens borskas AP blootstellings gebruik word. Agt verskillende blootstellings is afgelei met dosisse laer as die standaard. Die grootste afname was ongeveer 63%. Sewe van die blootstellings het ʼn beter visuele beeldkwaliteit as die standaard gehad, waarvan die grootste verbetering ongeveer 21% was. In twee van die agt beelde het ʼn analise van die kankerinduksierisiko getoon dat die risiko hoër as die standaard blootstelling kan wees, selfs al was die dosis laer. Bespreking en gevolgtrekking - Ses verskillende blootstellings is bepaal wat die dosis en kankerinduksierisiko verlaag het, met behoud of verbetering van beeldkwaliteit. Hierdie blootstellingsprotokolle word aanbeveel vir gebruik in die Diagnostiese Radiologie-afdeling in Tygerberg Akademiese Hospitaal.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/79971
This item appears in the following collections: