Design of a universal phatom for quality assurance in diagnostic radiology x-ray imaging

Annemari, Groenewald (2017-12)

Thesis (DSc)--Stellenbosch University, 2017.

Thesis

ENGLISH SUMMARY : Introduction: In medical X-ray imaging several diagnostic x-ray imaging modalities are applied to enable disease diagnosis, i.e. general projection radiography, fluoroscopy, mammography and Computed Tomography (CT) scanning. X-ray images must be of sufficient quality to enable accurate diagnosis. Image quality is quantified using suitable phantoms to ensure that equipment failure is detected before patient care is affected. A variety of phantoms are commercially available. However, these are modality specific, expensive and often complicated to use. In resource limited institutions, like many in Africa including South Africa, three problems are identified in the field of diagnostic radiology X-ray image quality control (QC). These are cost, man power and expertise and time constraints. A gap thus exists in the market for a single universal image quality assurance (QA) phantom, capable of doing all required QC tests for all X-ray imaging modalities. A phantom, answering to this requirement, in addition must be user-friendly and cost- and time-efficient. Aim: To design, develop, manufacture, test and validate a universal image QA phantom for diagnostic radiology X-ray imaging. The phantom must be compact, unique, universal (i.e. not modality specific), easy and quick to use and manufactured at a substantially reduced cost compared to the commercially available options. Materials and methods: Using literature studies on existing commercial phantoms for guidance, a prototype universal phantom was designed, manufactured and tested for all X-ray imaging modalities. From the prototype results, adjustments were made and the universal image quality phantom was developed and manufactured. The phantom is made from high density polyethylene and houses several inserts of different materials to assess sensitometry, image uniformity, limiting resolution, image noise, i.e. signal-to-noise (SNR) and contrast-to-noise (CNR) ratios, geometry and measurement tools, standard signal, low contrast detectability, positioning and alignment, artefacts and visual image quality inspection. For CT scanning the phantom measures slice thickness and for mammography masses, fibres and micro-calcifications are evaluated. Data analysis software was developed for analysis of obtained images and a complete step-by-step user’s manual was prepared. Reproducibility testing was performed on the phantom, using Department of Health (DoH) specified limits. Independent validation of the phantom package (i.e. phantom, software and manual) was done by three independent medical physicists. They compared the phantom to the commercial phantoms in general use in their institutes. Results: The universal image QA phantom and accompanying data analysis software produced reproducible results for all imaging modalities, within the accepted DoH tolerance levels. The independent validation results proofed that the phantom package was easy to transport, light weight and compact, easy to set-up and use, versatile, cost effective and user friendly. Discussion and conclusion: From the reproducibility testing and independent validation results it may be concluded that the universal image QA phantom, with accompanying data analysis software and user’s manual, offers an acceptable single phantom solution for medical X-ray imaging. The universal phantom is a cost and time saver and as such could fill a gap in the existing market. In addition, the phantom could also be used by radiographers in resource limited institutions.

AFRIKAANSE OPSOMMING : Inleiding: Mediese X-straalbeelding gebruik verskeie diagnostiese X-straal beeldingsmodaliteite om siekte te diagnoseer. Dít sluit algemene projeksieradiografie, fluoroskopie, mammografie en rekenaartomografie- (RT-)skandering in. Die X-straalbeelde moet van ’n voldoende gehalte wees om akkurate diagnose moontlik te maak. Beeldgehalte word met behulp van geskikte fantome gekwantifiseer om te sorg dat onklaar toerusting opgespoor word voordat dit pasiëntesorg beïnvloed. ’n Verskeidenheid fantome is kommersieel verkrygbaar. Tog is dit metodespesifiek, duur en dikwels ingewikkeld om te gebruik. By hulpbronbeperkte instellings, waaronder baie instellings in Afrika, wat Suid-Afrika insluit, word veral drie probleme met die gehaltebeheer van diagnostiese X-straalbeelding ondervind, naamlik koste, menslike hulpbronne en kundigheid, en tydsbeperkinge. Daar is dus ’n leemte in die mark vir ’n enkele, universele beeldgehalteversekeringsfantoom wat alle nodige gehaltebeheertoetse vir alle X-straalbeeldingsmetodes kan uitvoer. Daarbenewens moet so ’n fantoom gebruikersvriendelik en sowel koste- as tyddoeltreffend wees. Doelwit: Die doelwit is om ’n universele beeldgehalteversekeringsfantoom vir diagnostiese radiologie-X-straalbeelding te ontwerp, te ontwikkel, te vervaardig, te toets en te staaf. Die fantoom moet kompak, uniek en universeel (d.w.s. nie metodespesifiek nie) wees, sowel as maklik en vinnig om te gebruik. Boonop moet dit aansienlik goedkoper wees om te vervaardig as die huidige kommersieel verkrygbare fantome. Materiaal en metodes: Aan die hand van ’n literatuurstudie oor bestaande kommersiële fantome is ’n prototipe- universele fantoom vir alle X-straalbeeldingsmetodes ontwerp, gebou en getoets. Die ontwerp is op grond van die prototiperesultate aangepas, waarna die universele beeldgehaltefantoom ontwikkel en vervaardig is. Die fantoom word gemaak van hoëdigtheid-poliëtileen en bestaan uit verskeie invoegsels van verskillende materiale vir die evaluering van sensitometrie, beeldeenvormigheid, resolusiebeperking, beeldgeruis (met ander woorde sein-tot-geruis- en kontras-tot-geruis-verhoudings), geometrie en meetgereedskap, standaardsein, laekontrasopsporing, posisionering en belyning, artefakte, en visuele beeldgehalte. Vir RT-skandering meet die fantoom snitdikte, en vir mammografie word die opsporing van gewasse, vesels en mikroverkalkings geëvalueer. Dataontledingsagteware is ontwikkel om opgeneemde beelde te ontleed, en ’n volledige stapsgewyse gebruikershandleiding is saamgestel. Herhaalbaarheidstoetse is aan die hand van die Departement van Gesondheid se gespesifiseerde perke met die fantoom uitgevoer. Drie onafhanklike mediese fisici het die fantoompakket (d.w.s. fantoom, sagteware en handleiding) onafhanklik gestaaf. Hulle het die fantoom vergelyk met die kommersiële fantome wat hulle onderskeie instellings oor die algemeen gebruik. Resultate: Die universele beeldgehaltesekerheidsfantoom en gepaardgaande dataontledingsagteware lewer herhaalbare resultate vir alle beeldingsmetodes op, wat ook binne die Departement van Gesondheid se toleransieperke val. Die onafhanklike geldigheidsresultate bewys dat die fantoompakket maklik vervoer, liggewig en kompak is, maklik is om op te stel en te gebruik, en boonop veelsydig en kostedoeltreffend is. Bespreking en gevolgtrekking: Die herhaalbaarheidstoetse en onafhanklike geldigheidsresultate dui daarop dat die universele beeldgehaltesekerheidsfantoom, tesame met die gepaardgaande dataontledingsagteware en gebruikershandleiding, ’n aanvaarbare enkele fantoomoplossing vir mediese X-straalbeelding bied. Die universele fantoom spaar geld en tyd, en kan dus ’n leemte in die bestaande mark vul. Dit kan veral goed te pas kom vir radiografiste in hulpbronbeperkte instellings.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/102969
This item appears in the following collections: