A thesis submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree for the Doctor of Philosophy.
Johannesburg 2012 / Medical X-ray imaging is nowadays ubiquitous in healthcare. International studies have shown that patient doses during both diagnostic X-ray examinations and fluoroscopically guided procedures from one clinic to another can vary by a factor of up to 100. Such a variation in patient doses offers an opportunity for dose – image quality optimization. Given this background, every radiology clinic which wants to use X-ray imaging ethically and efficiently should have in place ways of optimizing the patient dose – image quality relationship. One generally accepted tool in the optimization process is diagnostic reference levels (DRLs). Currently in South Africa there are no established DRLs and there is no systematic patient dose data collection by the either the national regulator or any competent authority. The main purpose of this thesis was to quantify patient doses for patients undergoing diagnostic examinations and fluoroscopically guided procedures, educate radiation workers on typical patient doses, develop effective methods in quality control of radiographic and fluoroscopic equipment and evaluate radiographer familiarity with digital radiography technology within the context of a typical university teaching South African hospital. The present thesis comprises of seven studies, all carried out at Charlotte Maxeke Johannesburg Academic Hospital (CMJAH), formerly Johannesburg Hospital:
Study I: In this investigation the luminance level of X-ray viewing boxes and ambient lighting levels in reporting rooms were measured as a quality assurance procedure and compared with the recommended values by the Directorate of Radiation Control (DRC) of South Africa, European Commission (EC) and Nordic Radiation Protection Co-operation (NORDIC). Results from this investigation showed that the mean average luminance was 1027 cd m-2 and 3284 cd m-2 at the Division of Radiology and Division of Radiation Oncology respectively. The Division of Radiation Oncology had an average viewing box uniformity 7.14% compared to 27.32% at the Division of Radiology. The average ambient lighting was found to be 66 lux for both Divisions. The radiograph viewing conditions variably comply with guidelines. The radiographic imaging chain can only be as strong as its weakest link, thus this study underscores the need of implementing quality control and quality assurance standards in radiographic image viewing. Based on the practical experience of this investigation it is recommended that the DRC test criteria be adopted, in light of the varied recommendations worldwide.
Study II: This study aimed to develop, implement and evaluate a software program which can be used in a radiology quality control program. A Microsoft Excel™ based software program was developed for use in quality control: tests data collection, analysis and archiving of the tests done on general radiography equipment, fluoroscopy equipment and film processors. Validation of the software application in terms of usability, user-friendliness was done by an experienced radiographer. This software provides an easier and efficient way of recording quality control data, analysis and archiving.
Study III: This study retrospectively analyzed the radiation doses delivered to patients undergoing fluoroscopy guided procedures in terms of the skin dose1 and the kerma-area product readings. A total of three hundred and thirty one fluoroscopically guided procedures were analyzed. In agreement with other published studies, a weak correlation was shown between skin dose and screening time, while a poor correlation was shown between KAP reading and screening time. There was a wide spread in the radiation doses registered for any one given type of examination, which shows that there is room for dose optimization. From the lessons drawn from this study it is practically feasible to record the KAP, fluoroscopy time and number of images routinely. The usefulness and potential use of KAP meters with regards to dose optimization in radiology was confirmed.
Study IV: This investigation aimed to assess the feasibility of fabricating in-house the clinical dosimetry radiology phantoms. A total of six patient dose assessment phantoms were fabricated of which four phantoms were as per American National Standards Institute (ANSI) specifications and the other two as per Centre for Devices and Radiological Health (CDRH) specifications. This study proved that the phantoms can be fabricated cost-effectively in-house in a hospital with a mechanical engineering workshop using materials which are locally available. In addition, this study determined radiation doses received by patients undergoing six general radiography examinations. The feasibility of both direct and indirect methods of patient dosimetry was studied. Patient dosimetry based on indirect measurements was the method of choice. Patient data and technical parameters related to the X-ray examinations were collected. The study involved the following examinations: chest posterior-anterior (PA), chest lateral (LAT), pelvis anterior-posterior (AP), abdomen AP, lumbar spine AP and thoracic spine AP. Entrance surface air kerma was calculated based on the X-ray tube
1 See Section 8.3 on the use of the term skin dose
output of the unit used and the exposure parameters used for the actual examination. Based on the mean entrance surface air kerma (ESAK) values from the individual rooms, the following DRLs were established: 0.10 mGy for chest PA, 0.22 mGy for chest LAT, 2.98 mGy for pelvis AP, 4.19 mGy for abdomen AP, 5.30 mGy for lumbar spine AP and 3.28 mGy for thoracic spine AP. The calculated mean ESAK values were compared with previously published mean values from other countries. For the first time, a baseline for potential dose reference levels (DRLs) in South Africa was established for the selected examinations. The results of this snapshot audit serve as a benchmark for future dose optimization attempts in South Africa. Feasible and practical dose saving measures are presented and discussed based on the experience of the present patient dose audit carried out.
Study V: A replica of the CDRAD phantom was successfully fabricated in-house for use as an image quality test object. It has been shown that the phantom when fabricated in-house is inexpensive and can be made from materials that are readily available locally. Furthermore the utility of the replica phantom as both an acceptance testing and routine quality control tool has been demonstrated. The replica phantom proved effective for purpose and user-friendly.
Study VI: The purpose of this study was to assess radiographer familiarity and preferences with digital radiography and thereafter make recommendations in line with the migration from screen film to digital radiography in South Africa. A questionnaire was designed to collect data from either qualified or student radiographers from four teaching hospitals. From the four teaching hospitals there were a total of 205 potential respondents. Among other things, responses regarding experiences and preferences with digital radiography, quality control procedures, patient dose, advantages and disadvantages of digital radiography were sought. The information collected was based on self-reporting by the participants. Sixty-three out of 205 (31 %) radiographers from all the four radiology centres responded to the circulated questionnaire. The participants of this survey showed familiarity with digital radiography and have embraced this relatively new technology as shown by the fact that they can identify both its advantages and disadvantages as applied to clinical practice. However, there are minimal quality control procedures specific to digital radiography being undertaken and there is need for formal education, continuing education and manufacturer training with
respect to quality control as institutions make the transition from conventional screen film radiology to digital radiology.
Study VII: An investigation into the amount of scattered radiation from the couch during under-couch procedures was carried out. Of dosimetric concern are the forward scattered photons from the couch which contribute in principle to patient dose. Measurement of the amount of scattered radiation off the patient couch was accomplished by using an ionization chamber. The results of the investigation showed that for field size of dimensions, 10 cm * 10 cm, the scatter contribution is approximately 12 % of the total radiation reaching the patient surface. In addition the scatter contribution varies by ±2% across field sizes ranging from 8 cm * 8 cm to 20 cm * 20 cm, with the 10 cm * 10 cm field size taken as a reference field. This study underscores the need to account for the forward scattered radiation so as to improve the accuracy of clinical patient dosimetry.
Programs of continuing education and training of radiological personnel in appropriate radiological technique need be actively implemented in order to maintain a high level of awareness of the factors that determine the diagnostic quality and dose to the patients. In line with efforts to optimize dose from diagnostic radiography examinations it is recommended that national DRLs be established in South Africa for the most frequent examinations in general radiography and fluoroscopy. It is recommended that the South African national regulator endeavour to implement or facilitate implementation of a national patient dose database. In summary, this thesis indicates the possibility of dose reduction in diagnostic radiology through optimization of radiographic process.
|02 May 2013
|South African National ETD Portal
Page generated in 0.0032 seconds