Date of Award
5-2014
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Legacy Department
Bioengineering
Committee Chair/Advisor
Hai Yao, PhD
Committee Member
Walter Huda, PhD
Committee Member
Richard Swaja, PhD
Committee Member
Tong Ye, PhD
Abstract
Radiation doses are caused by the energy deposited in unit mass of matter from ionizing radiation. In the US, radiation doses from medical imaging increased six-fold in the past generation. Among medical exposures to patients, computed tomography (CT) composes about half of the collective doses, and interventional fluoroscopy composes 14%. Radiation exposure to patients undergoing diagnostic radiological procedures causes increased lifetime carcinogenic risks, especially for pediatric patients who are more radiosensitive than adults. The correlation between procedural x-ray techniques and the radiation doses to patients, as well as the resultant image quality, is not well understood, and therefore the focus of the performed studies. High radiation dose levels can occur as an outcome of complex procedures requiring additional imaging, or when a patient undergoes multiple radiological procedures. Accumulated occupational doses, caused by the scattered radiation from the patient to the staff during the procedures, are also of concern. There are many factors that affect the patient radiation doses, such as different combinations of technical parameter settings and patient characteristics. Due to the complexities and time-consuming nature of clinical dose/exposure measurements, the Monte Carlo technique is the only realistic tool to investigate patient doses and occupational exposure. Therefore, the objective of this Dissertation is to investigate the possible optimization methods of the irradiation technical factors in order to lower radiation doses to patients undergoing diagnostic radiological examinations using Monte Carlo algorithm-based software. Our general hypothesis is that incident x-ray photon energy used in a diagnostic radiological procedure can be optimized to reduce patient doses without sacrificing image quality, and therefore can lower radiation-induced lifetime carcinogenic risks for patients. Our results will be valuable for medical physicists to analyze dose distributions, and for the cardiology clinicians to maximize image guidance capabilities while minimizing potential carcinogenic and deterministic risks to pediatric patients. Firstly, the impact of irradiation parameters on patient doses during CT scans was investigated and possible optimization methods were discussed. Our results about cone beam CT scans showed that there were major differences in organ and effective dose as the x-ray tube rotates around the patient. This suggested that the use of x-ray tube current modulation could produce substantial reductions in organ and effective dose for body imaging with cone beam CT. For chest CT, our results showed that the existing x-ray tube current modulation schemes are expected to reduce patient effective doses in chest CT examinations by about 10%, with longitudinal modulation accounting for two thirds and angular modulation for the remaining one third. It was also shown that the choice of the scanned region affects organ doses in CT. Secondly, the radiation-induced cancer risks from body CT examinations for adult patients were estimated. For patients who differ from a standard sized adult, correction factors based on the patient weight and antero-posterior dimension are provided to adjust organ doses and the corresponding risks. Our results showed that at constant incident radiation intensity, for CT examinations that include the chest, risks in females are markedly higher than those for males, whereas for examinations that include the pelvis, risks in males were slightly higher than those in females. In abdominal CT scans, risks for males and female patients are very similar. A conclusion was reached that cancer risks in body CT can be estimated from the examination Dose Length Product by accounting for sex, age, as well as patient physical characteristics. Thirdly, a set of innovative Monte Carlo models were developed to investigate the role of x-ray photon energy in determining skin dose, energy imparted, and image quality in pediatric interventional radiology using the MCNP5 platform. Contrast, relative noise, and contrast-to-noise ratio (CNR) were obtained for diagnostic imaging with and without the utilization of grids. Our results indicated that using Monte Carlo methods, the optimized x-ray tube voltage for a relatively low patient dose under the desired image quality could be obtained for any specific patient undergoing a certain type of diagnostic examination. Lastly, we investigated the changes in the pattern of energy deposition in patient phantoms following the use of iodinated contrast media using Monte Carlo models built on MCNP5 platform. Relative energy imparted to the volume of interest with iodine contrast agent, as well as to the whole patient phantom, was calculated. Changes in patterns of energy deposition around the contrast-filled volume were also investigated. Our results suggested that adding iodine can result in values of localized absorbed dose increasing by more than an order of magnitude, but the total energy deposition is generally very modest. Furthermore, our results also showed that adding iodine primarily changes the pattern of energy deposition in the irradiated region, rather than increasing the corresponding patient doses. The goal of this project was to establish a better understanding of the roles of different technique factors in the patient doses from diagnostic radiological procedures. Based on these studies, the limitations of the current Monte Carlo software were analyzed and our own Monte Carlo model was proposed for simulations of patient doses during pediatric interventional radiology procedures. The ultimate goal of this study is to develop a comprehensive dosimetry database using Monte Carlo technique, with the output of patient doses, operator doses, and the corresponding radiation-induced carcinogenesis risks for pediatric interventional radiology procedures.
Recommended Citation
He, Wenjun, "MONTE CARLO MODELING BASED PATIENT DOSE OPTIMIZATION IN DIAGNOSTIC RADIOLOGY" (2014). All Dissertations. 1276.
https://open.clemson.edu/all_dissertations/1276