Standardization and calibration for breast MRI : the use of a unique air bubble free agar phantom

M. Heisen, A.M. Wood, J. Buurman, D. Mustafi, E. Peng, G.M. Newstead, C. Yang, M.K. Ivancevic, G.S. Karczmar

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Downloads (Pure)


PURPOSE The routine use of an in vivo phantom could provide the basis for standardization and calibration of DCE-MRI of the breast. Here, it is used to correct for imperfect flip angle profiles to obtain a more accurate T1 quantitation. METHOD AND MATERIALS A unique soft phantom was developed for clinical use. The air bubble-free ultra-pure agar content of the ring-shaped phantom covers a range of contrast agent concentrations (Omniscan, 0 – 1.0 mM). As a reference, an accurate variable flip angle measurement (3D FSPGRE, FA= 3/5/10/15/20/25/30/35/40/50°, TR/TE=25/1.92 ms, NA = 4) was performed on the phantom. A healthy volunteer (no contrast, IRB approval) was scanned in a standard bilateral coil at 1.5 T, with the phantom placed around the breast. Again a variable flip angle measurement was performed (3D FSPGRE, FA = 5/10/20/25/40°, TR/TE=25/1.9 ms, NA = 1). Based on the signal intensities and the previously measured T1 in the phantom, the effective flip angles were simultaneously estimated and used to calculate T1 for the breast. In addition, to ensure that the presence of the phantom does not interfere with the appearance of the breast, clinical T2-w (2D turbo spin echo) and T1-w (ultrafast gradient echo) images were acquired with and without the phantom around the breast. RESULTS Under the assumption of a uniform B1 field of the body coil in the central slices, the effective flip angles were estimated at: 4.8/9.8/21.9/27.3/57.1°, resulting in corrected T1 estimates for fat and parenchyma of 261.4 ± 6.6 ms (without flip angle correction: 344.5 ± 54.3 ms), and 763.0 ± 31.3 ms (without correction: 930.1 ± 154.1 ms), respectively. The quality of the clinical images containing the phantom was found to be adequate by an experienced breast radiologist. Its presence did, however, affect automatic windowing. No discomfort was reported. CONCLUSION This in vivo phantom can be used to improve T1 quantitation. Alternatively, it may be used to quantify T1 without the need for a variable flip angle measurement. A reliable phantom could also improve quantitative measurements of proton density. An improved phantom design is under development and will be used in routine clinical practice. CLINICAL RELEVANCE/APPLICATION The routine use of a reliable clinical calibration phantom could lead to improved standardization of DCE-MRI across different institutions.
Original languageEnglish
Title of host publicationProceedings of the RSNA 2009, November 29-December 4, 2009, Chicago
Place of PublicationChicago
Publication statusPublished - 2009


Dive into the research topics of 'Standardization and calibration for breast MRI : the use of a unique air bubble free agar phantom'. Together they form a unique fingerprint.

Cite this