The Implications of Patient Centring on CT Number and Radiation Dose in CT Imaging

Background:
Since the advent of helical scanning and multidetector-row technology, computed tomography (CT) has significantly improved volume coverage, acquisition speed, and longitudinal resolution. These advancements have consequently enhanced the provision of detailed anatomical information through three dimensional (3D) reconstructed images and led to remarkable improvements in image quality, with the accuracy of CT numbers or Hounsfield Unit (HU) being one of the key metrics characterising this quality. This accuracy is essential for various clinical applications to ensure reliable diagnostic and therapeutic outcomes.

However, the increasing use of CT examinations worldwide has led to increased radiation exposure to patients, raising concerns about potential cancer risks. To mitigate these risks, various radiation dose reduction strategies have been developed, including automatic tube current modulation (ATCM) and beam-shaping filters. These strategies require precise patient positioning at the gantry iso-centre, the central point within the gantry aperture where the gantry's axis of rotation intersects, to function optimally. Failure to achieve accurate positioning (centring) can detrimentally affect both the radiation dose and image quality. Therefore, CT radiographers must prioritise meticulous patient positioning to ensure optimal imaging outcomes.

Objectives:
This thesis aims to investigate the impact of incorrect patient centring on radiation dose and the accuracy of CT numbers during multidetector computed tomography (MDCT) imaging using different phantoms. This investigation considered several factors: tube voltage, patient size and anatomical location, scanner manufacturer model, localiser direction, body position (supine or prone), and bowtie filter selection.

The research provides a comprehensive analysis of the operational principles of dose optimisation tools and the significant influence of patient centring on their optimal functionality. The ultimate goal is to enhance the efficiency of dose reduction technologies, thereby minimising radiation exposure to patients and healthcare professionals. The results of this research, grounded in empirical evidence, will be invaluable in formulating guidelines that enhance patient safety and the quality of medical imaging. Furthermore, this thesis will evaluate the reliability of CT numbers, which are essential for various clinical applications, such as characterising tissue lesions in diagnostic radiography, assessing bone mineralisation for implant site planning using cone beam CT (CBCT) in dentomaxillofacial imaging, and radiation therapy treatment planning.

Finally, this thesis intends to report the current practices and accuracy knowledge of radiographers in Australia concerning the impact of off-centring, localiser radiograph selection, scanning parameters, patient positioning, and patient size on radiation dose and CT number accuracy. The findings of this research will provide valuable insights that can be used to promote continuing professional development (CPD) and training focused on best-practice positioning.

Methods and Materials:
This thesis comprises three experimental studies and one survey. The studies included are titled:
Study 1: The Impacts of Vertical Off-Centring, Tube Voltage, and Phantom Size on Computed Tomography Numbers: An Experimental Study

Study 2: The Impacts of Vertical Off-Centring, Localizer Direction, Phantom Positioning and Tube Voltage on CT Number Accuracy: An Experimental Study

Study 3: 0⁰ vs. 180⁰ CT Localizer: The Effect of Vertical Off-Centring, Phantom Positioning and Tube Voltage on Dose Optimization in Multidetector Computed Tomography

Study 4: The Effect of Inappropriate Patient Centring on CT Numbers and Radiation Dose: A Survey of Current Practices and Knowledge

Study 1 utilised CIRS Model 062 Electron Density and system performance phantoms imaged on Siemens Emotion 16-slice CT and GEMINI-GXL scanners, respectively. The uniformity and accuracy of CT numbers were assessed as a function of vertical off-centring by employing various water phantom sizes (18, 20, and 30 cm) and tube voltages (80, 90, 110, 120, 130, and 140 kVp). Paired t-tests were used to compare the changes in CT numbers between the two different scanners and the vertical off-centring. The significance of the impact of phantom off-centring and body size on the CT number was evaluated using analysis of covariance (ANCOVA).

Studies 2 and 3 involved scanning the torso of a PBU-60 anthropomorphic phantom (Kyoto Kagaku Co., Ltd., Kyoto, Japan) that represents an average-sized adult at six vertical levels using a Discovery CT750 HD - 128 slice (GE Healthcare) MDCT scanner. Images were obtained at various combinations of vertical off-centring (above and below the gantry iso-centre), localiser direction (0° and 180°), phantom position (supine or prone), and energy level applied (80, 120, and 140 kVp). The apparent phantom size (projected localiser size) was determined for each vertical height position as the projected width from the localiser radiograph. The volume CT dose index (CTDIvol), dose length product (DLP), and CT numbers in the upper thorax, mid-thorax, and abdomen were recorded for each vertical off-centring increment and experimental configuration. A dependent paired t-test was conducted to explore the dose as a function of vertical off-centring between phantom positions (supine and prone) using the Statistical Package for the Social Science Version 26 (SPSS). Pearson’s correlation coefficient was used to assess the relationship between the apparent phantom size and vertical off-centring. Furthermore, a one-way analysis of variance (ANOVA) was performed to measure the statistical differences across phantom vertical off-centring, scanned region, and tube voltage on CT numbers. A paired-sample t-test was used to determine the statistical difference in CT numbers between the phantom positions (supine vs. prone) and between the localisers’ directions (0° vs. 180°) as a function of vertical off-centring. Pearson’s correlation test was used to assess the relationship between radiation dose (CTDIvol) and CT number change as a function of phantom vertical off-centring at different localiser directions and phantom positions.

Study 4 reported, for the first time, the current practice and accuracy of knowledge of radiographers in Australia regarding patient positioning, radiation dose, and CT number accuracy during CT examinations. An online survey was created and directed towards diagnostic radiographers registered with the Australian Health Practitioner Regulation
Agency (Ahpra), and members of the Australian Society of Medical Imaging and Radiation Therapy (ASMIRT). The eligibility requirements included clinical proficiency in CT imaging through clinical practice and/or academic qualifications. The survey invitation was disseminated through the ASMIRT online newsletter (eBlast) to provide broad geographic coverage for participants who met the inclusion criteria within a short timeframe. The survey underwent a pilot test involving three qualified CT radiographers, and minor adjustments were made, including the rephrasing of specific questions and the use of alternate terminology. The survey comprised 36 questions, consisting of 19 multiple-choice and 17 Likert-scale questions, with five response options ranging from “strongly disagree” to “strongly agree”, and “never” to “always”. Ethical approval was obtained from the Charles Sturt University Human Research Ethics Committee (HREC), approval number H21064. Chi-square analysis was used to assess statistical significance for nominal data, and Student's t-test for continuous data. Pearson Chi-Square (X2) test was used; if abnormally distributed, the Likelihood Ratio Chi-Square (G2) test was used. An ANOVA was used (grouped F-test) to analyse of variance among three or more variables. Differences were considered statistically significant at a p-value < 0.05.

All CT scanners used in this study were clinically operational and subject to regular calibration as per national legislation. Calibration is conducted during the commissioning of each scanner and as part of routine quality assurance (QA) processes, including daily, monthly, quarterly, and annual checks. These procedures are essential to ensure the accurate performance of the scanners, thereby maintaining the reliability and consistency of the CT imaging data.

Results:
The results of Study 1 revealed a significant difference in the uniformity of the CT numbers between the scanners. It was found that vertical off-centring and phantom size accounted for 92% of the observed variability in CT number change, attributed to alignment between the scanned phantom size and the applied bowtie filter. Both systems demonstrated a more pronounced influence of vertical off-centring on CT numbers at lower tube voltages. The maximum absolute difference in CT number for water between two oppositely located regions of interest (ROIs) was notably 13.7 HU with 80 mm vertical off-centring.

In study 2, vertical off-centring affected the accuracy of CT numbers. The variation was primarily affected by the tube voltage, location of the measured ROI, phantom orientation (supine or prone), localiser direction, and the anatomical scanned region. The maximum observed change in soft tissue was 43 HU. Importantly, a strong positive correlation was recorded between the dose variation and CT numbers as a function of vertical off-centring.

Study 3 further demonstrated that vertical off-centring, localiser direction, and phantom position significantly impacted the radiation dose when ATCM was utilised. The dose variation associated with vertical off-centring depended on the apparent localiser size and orientation of the phantom (supine or prone). Shifting the phantom toward the X-ray tube magnifies the localiser image, resulting in a higher radiation output (photon flux) with a maximum increase of 196% compared to the dose at the gantry iso-centre.

Study 4, which involved a survey of Australian radiographers, identified inconsistencies in CT practices and knowledge gaps concerning the impact of precise patient positioning during CT examinations and its implications on dose and CT numbers. These findings were influenced by various factors, including CT qualifications, workload, and years of experience. Although almost all participants recognised the importance of CT number accuracy in prognosis and clinical decisions, approximately half rarely reviewed the CT number after image acquisition. A noteworthy finding was that only a minority acknowledged the impact of the localiser direction on the radiation dose when patients were improperly centred. Further, adherence to manufacturers’ recommendations regarding localiser selection was noted, regardless of the potential consequences on dose and image quality if patients were not accurately positioned. In other words, despite radiographers' awareness of potential patient off-centring, such as due to patient size or medical equipment attached to the patient, they consistently follow the manufacturer's recommendations for localiser selection. However, adjusting the localiser direction based on the off-centring could significantly influence both radiation dose and image quality, a consideration that was not widely recognised among survey participants.

Conclusion:
These findings underscore the critical role of patient positioning in optimising the radiation dose and enhancing CT number accuracy during CT examinations. Off-centring a scanned object closer to the X-ray tube resulted in an enlarged localiser image, leading to an increased radiation output facilitated by the ATCM system's higher photon flux. The extent of CT number variability resulting from off-centring was influenced by the improper alignment of the bowtie filter with the phantom, particularly by impacting larger phantoms and lower tube voltages. Factors such as localiser direction, phantom position, scanned region, and location of the ROI were identified as significant variables affecting the degree of CT number variation. The identified inconsistencies in practices and knowledge gaps among Australian radiographers emphasise the importance of CPD activities to stay up-to-date with the evolving best practice evidence base and technological advancements in the field, including advancements in auto-positioning aiding software.