Click here to


Are you sure ?

Yes, do it No, cancel

Radiological and Dosimetric Characterization of a CT, MRI and Ultrasound Customizable Breast Phantom for Radiotherapy Use and Surgical Practice

G Aldosary1*, T Tse2 , W Lee3 , T Alotaibi4 , C Foottit5,6 , A Arnaout2,7 , J Caudrelier3,8 , E Vandervoort5,6 , J Belec5,6 , (1) Carleton University, Ottawa, Canada (2) General Surgery, The Ottawa Hospital, Ottawa, Canada (3) Radiation Medicine, The Ottawa Hospital, Ottawa, Canada (4) Medical Imaging, The Ottawa Hospital, Ottawa, Canada (5) Department of Radiology, Faculty of Medicine, The University of Ottawa, Ottawa, Canada (6) Medical Physics, The Ottawa Hospital, Ottawa, Canada (7) Department of Surgery, The University of Ottawa, Ottawa, Canada (8) Department of Medicine, The University of Ottawa, Ottawa, Canada


(Wednesday, 7/17/2019) 10:30 AM - 11:00 AM

Room: Exhibit Hall | Forum 6

Purpose: The displacement of tumor bed walls and surgical clips during oncoplasty breast surgery (OBS) decreases the clips’ reliability as radiographic surrogates of tumor bed locations—potentially compromising radiotherapy target delineation. A breast phantom was employed to investigate the clips’ reliability and to facilitate OBS practice. This work focuses on the radiological and dosimetric characteristics of this phantom.

Methods: A realistic silicone-based breast phantom was constructed with composition informed by two experienced breast surgeons. It contained materials emulating a breast parenchyma, epidermis, areola, nipple, chest wall and 8 titanium surgical clips. The phantom was imaged with CT, MRI (T1w- & T2w-images) and ultrasound. Using ImageJ, the parenchyma’s signal-to-noise ratio (SNR) was measured. The contrast-to-noise (CNR) and SNR were also measured for clips vs. parenchyma on CT and ultrasound. The CT was imported to the Monaco TPS, and the mean relative electron density (RED) and Hounsfield Units (HU) were measured. The RED was compared to measured patient data (sample n=5). The calculated dose in phantom at central-axis depth= 3.2 cm was compared with radiochromic film measurements (field size=4x4 cm², MU=300, SSD=100 cm, 6 MV).

Results: The breast phantom components were visible on all imaging modalities. The mean RED=1.06 ± 0.01 (patient data mean= 0.90 ± 0.02). Parenchyma image SNR= 13, 46, 44 & 27 on CT, T1w-, T2w-MRIs and ultrasound, respectively. For clips, the CNR= 193 and 36, and SNR= 205 and 49, for CT and ultrasound images, respectively. Compared to the calculated mean dose, film measurements showed mean dose differences of +1.0%.

Conclusion: A phantom was constructed to mimic breast tissue by palpation and on multi-imaging modalities. The phantom’s imaging characteristics suggest its resemblance to a fibroglandular breast and allow clear visualization of high-density markers (e.g. surgical clips/fiducials) used in radiotherapy. Dosimetric results confirmed the phantom’s appropriateness for radiotherapy applications.


Phantoms, Breast, MRI


TH- Radiation dose measurement devices: Phantoms for dosimetric measurement

Contact Email