Back to Search
Start Over
Low-Dose Contrast-Enhanced Breast CT Using Spectral Shaping Filters: An Experimental Study
- Source :
- IEEE Transactions on Medical Imaging. 36:2417-2423
- Publication Year :
- 2017
- Publisher :
- Institute of Electrical and Electronics Engineers (IEEE), 2017.
-
Abstract
- Iodinated contrast-enhanced X-ray imaging of the breast has been studied with various modalities, including full-field digital mammography (FFDM), digital breast tomosynthesis (DBT), and dedicated breast CT. Contrast imaging with breast CT has a number of advantages over FFDM and DBT, including the lack of breast compression, and generation of fully isotropic 3-D reconstructions. Nonetheless, for breast CT to be considered as a viable tool for routine clinical use, it would be desirable to reduce radiation dose. One approach for dose reduction in breast CT is spectral shaping using X-ray filters. In this paper, two high atomic number filter materials are studied, namely, gadolinium (Gd) and erbium (Er), and compared with Al and Cu filters currently used in breast CT systems. Task-based performance is assessed by imaging a cylindrical poly(methyl methacrylate) phantom with iodine inserts on a benchtop breast CT system that emulates clinical breast CT. To evaluate detectability, a channelized hoteling observer (CHO) is used with sums of Laguerre-Gauss channels. It was observed that spectral shaping using Er and Gd filters substantially increased the dose efficiency (defined as signal-to-noise ratio of the CHO divided by mean glandular dose) as compared with kilovolt peak and filter settings used in commercial and prototype breast CT systems. These experimental phantom study results are encouraging for reducing dose of breast CT, however, further evaluation involving patients is needed.
- Subjects :
- Digital mammography
Gadolinium
Contrast Media
chemistry.chemical_element
Breast Neoplasms
Iterative reconstruction
Radiation Dosage
Imaging phantom
030218 nuclear medicine & medical imaging
Spectral shaping
03 medical and health sciences
0302 clinical medicine
Image Interpretation, Computer-Assisted
Humans
Medicine
Breast
Electrical and Electronic Engineering
skin and connective tissue diseases
Breast ct
Radiological and Ultrasound Technology
Phantoms, Imaging
business.industry
Low dose
Computer Science Applications
chemistry
030220 oncology & carcinogenesis
Female
Tomography
Tomography, X-Ray Computed
business
Nuclear medicine
Algorithms
Software
Subjects
Details
- ISSN :
- 1558254X and 02780062
- Volume :
- 36
- Database :
- OpenAIRE
- Journal :
- IEEE Transactions on Medical Imaging
- Accession number :
- edsair.doi.dedup.....03cd789f68e72d7bb13ad9a8fddb7b62