Characterization of a Solid-State Detector for Dosimetry in the Diagnostic Energy Range with Verification via Monte Carlo Estimation of Average Breast Dose in Mammography
Doctor of Philosophy
Pei-Jan Paul Lin, Ph.D.
Monica Ghita, Ph.D.
William Song, Ph.D.
Priti Shah, M.D.
Mariela A. Porras-Chaverri, Ph.D.
This dissertation aims to develop a simplified QA metric for estimating radiation dose in mammography. This metric, the Mid-Breast Dose (MDB) Index, maybe be proposed as an alternative dose index, consistent with average glandular dose (AGD), for routine QA. MBD was obtained using “Phantom mid-point Air Kerma (AK) measurement” for conventional mammography. The advantage of this method is that MBD can be measured directly without requiring multiple conversion factors. The accuracy of clinical measurements in computing the MBD was assessed by comparing it with the AGD estimated by the Dance formalism experimentally and Monte Carlo simulation.
MBD methodology relies on detector design. Detectors must be thin for insertion between phantom layers. Therefore, MBD measurements were obtained using ionization chamber and two solid-state detectors (SSDs). One of the SSDs was modified to include backscatter. MBD metric was evaluated experimentally using an equivalent breast tissue BR12 phantom with a composition of 50% glandular tissue and 50% adipose tissue.
The results show the MBD using “phantom mid-point AK measurement” method with an either ionization chamber or a properly calibrated SSD provides comparable results with the ACR methodology for the AGD. The MBD is not a patient-related dose. It is a simplified quality assurance dose metric for (4.2 cm) compressed breast thickness with 50% adipose and 50% glandular tissue. This metric can be proposed as a suitable dose index for routine QA. With this method, the mid-point AK in phantoms and one unique conversion factor (0.72), are required across all beam qualities in mammography.
© Areej Aljabal
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
Available for download on Friday, May 10, 2024