X-ray phase contrast imaging of the breast: Analysis of tissue simulating materials

Purpose: Phase contrast imaging, particularly of the breast, is being actively investigated. The purpose of this work is to investigate the x-ray phase contrast properties of breast tissues and commonly used breast tissue substitutes or phantom materials with an aim of determining the phantom materials best representative of breast tissues. Methods: Elemental compositions of breast tissues including adipose, fibroglandular, and skin were used to determine the refractive index, n = 1 - delta + i beta. The real part of the refractive index, specifically the refractive index decrement (delta), over the energy range of 5-50 keV were determined using XOP software (version 2.3, European Synchrotron Radiation Facility, France). Calcium oxalate and calcium hydroxyapatite were considered to represent the material compositions of microcalcifications in vivo. Nineteen tissue substitutes were considered as possible candidates to represent adipose tissue, fibroglandular tissue and skin, and four phantom materials were considered as possible candidates to represent microcalcifications. For each material, either the molecular formula, if available, or the elemental composition based on weight fraction, was used to determine delta. At each x-ray photon energy, the absolute percent difference in delta between the breast tissue and the substitute material was determined, from which three candidates were selected. From these candidate tissue substitutes, the material that minimized the absolute percent difference in linear attenuation coefficient mu, and hence delta, was considered to be best representative of that breast tissue. Results: Over the energy range of 5-50 keV, while the delta of CB3 and fibroglandular tissue-equivalent material were within 1% of that of fibroglandular tissue, the mu of fibroglandular tissue-equivalent material better approximated the fibroglandular tissue. While the delta of BR10 and adipose tissue-equivalent material were within 1% of that of adipose tissue, the tissue-equivalent material better approximated the adipose tissue in terms of mu. Polymethyl methacrylate, a commonly used tissue substitute, exhibited delta greater than fibroglandular tissue by similar to 12%. The A-150 plastic closely approximated the skin. Several materials exhibited delta between that of adipose and fibroglandular tissue. However, there was an energy-dependent mismatch in terms of equivalent fibroglandular weight fraction between delta and mu for these materials. For microcalcifications, aluminum and calcium carbonate were observed to straddle the delta and mu of calcium oxalate and calcium hydroxyapatite. Aluminum oxide, commonly used to represent microcalcifications in the American College of Radiology recommended phantoms for accreditation exhibited delta greater than calcium hydroxyapatite by similar to 23%. Conclusions: A breast phantom comprising A-150 plastic to represent the skin, commercially available adipose and fibroglandular tissue-equivalent formulations to represent adipose and fibroglandular tissue, respectively, was found to be best suited for x-ray phase-sensitive imaging of the breast. Calcium carbonate or aluminum can be used to represent microcalcifications. (C) 2013 American Association of Physicists in Medicine. [http://dx.doi.org/10.1118/1.4794503]