Radiation dose, cancer risk and diagnostic performance of computed tomography pulmonary angiography examination

Concerns towards high radiation dose and cancer risk from Computed Tomography Pulmonary Angiography (CTPA) examinations have prompted efforts to develop a novel optimization while preserving CT image diagnostic performance. Hence, this study aims to evaluate the radiation dose, cancer risk and image diagnostic performance of CTPA examination regarding primary and secondary optimization such as iterative reconstruction (IR) algorithm, tube potential and pitch factor selection. The first phase of this thesis begins with the establishment of a local Diagnostic Reference Levels (DRL) with respect to image quality together with evaluation of organ dose and cancer risk. 127 subjects (55 men and 72 women) with an age range from 18 to 88 years old who were suspected of having PE and underwent CTPA examination were recruited. Dose descriptors such as volume-weighted CT Dose Index (CTDIvol), Size- Specific Dose Estimates (SSDE), Dose Length Product (DLP) and effective dose (E) were recorded and analyzed together with noise as the image quality. Body sizes of the subjects were categorized based on their effective diameter (ED) length and divided into three groups: P1 (19–24 cm), P2 (24–29 cm), and P3 (29–34) cm. There is a significant difference in local DRL values and between body sizes (p < 0.05) while noise is not significantly different between body sizes (p > 0.05). Organ dose and cancer risk were estimated by the CT-EXPO (Ver 2.5.1, Germany) and recommendation from the International Commission on Radiological Protection Publication (ICRP) 103 report respectively according to the primary beam. In one million CTPA examinations, the risk of cancer in breast, lung and liver organs was 0.009%, 0.007%, and 0.005%, respectively. The second phase focuses on the diagnostic performance assessment. Different levels of IR algorithms were applied in routine CTPA and several modified tube potentials. The value of signal and noise were defined by placing the circular region of interest (ROI) on the main pulmonary artery (MPA), right pulmonary artery (RPA), left pulmonary artery (LPA), ascending aorta (AA), and descending aorta (DA). The performance of each protocol was presented as Signal to Noise Ratio (SNR), Contrast to Noise Ratio (CNR) and Figure of Merit (FOM). CNR and FOM performed significantly better when IR algorithm levels were increased, tube potential was reduced, and if patients had smaller body sizes (p < 0.05) while fluctuation trends were observed in SNR. The last phase of this thesis covers the assessment of diagnostic performance by a CATPHAN 600 phantom with primary and secondary optimization. The phantom was scanned with CTPA local protocol using different tube potentials and pitch factors. Images obtained were reconstructed with the IR algorithm (levels 3, 4 and 5). Diagnostic performance was quantified objectively by imQuest software (version 7.1, Duke University, USA). Noise power spectrum (NPS), target transfer function (TTF), CNR and SNR were evaluated concerning the optimization setting. It was found that the CNR value was increased while NPS was degraded by increasing IR levels. Noise value reduction has significantly achieved the increase of tube potential although there are no changes in TTF values. The alteration of pitch factor provides some fluctuation pattern of both NPS and TTF values. Future research could be carried out with different types of examinations, scanner models and substantive phantoms to expand these assessment methods proposed in this study. It is also recommended to extend a subjective diagnostic performance measurement to increase the reliability of the objective measurement attained. The study discovered a novel finding of the characterization of radiation dose, cancer risk and diagnostic performance with a different type of quantitative measurement for local CTPA examination.

Text FS 2021 28 IR.pdf Download (1MB)

Actions (login required)

View Item