Premium
Detecting breast microcalcifications with high‐field MRI
Author(s) -
Leeuw Hendrik,
Stehouwer Bertine L.,
Bakker Chris J. G.,
Klomp Dennis W. J.,
Diest Paul J.,
Luijten Peter R.,
Seevinck Peter R.,
Bosch Maurice A. A. J.,
Viergever Max A.,
Veldhuis Wouter B.
Publication year - 2014
Publication title -
nmr in biomedicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.278
H-Index - 114
eISSN - 1099-1492
pISSN - 0952-3480
DOI - 10.1002/nbm.3089
Subject(s) - mammography , breast imaging , magnetic resonance imaging , nuclear medicine , scanner , microcalcification , digital mammography , radiology , nuclear magnetic resonance , computer science , artificial intelligence , medicine , breast cancer , physics , cancer
The aim of this study was to detect microcalcifications in human whole breast specimens using high‐field MRI. Four mastectomy specimens, obtained with approval of the institutional review board, were subjected to gradient‐echo MRI acquisitions on a high‐field MR scanner. The phase derivative was used to detect microcalcifications. The echo time and imaging resolution were varied to study the sensitivity of the proposed method. Computed tomography images of the mastectomy specimens and prior acquired mammography images were used to validate the results. A template matching algorithm was designed to detect microcalcifications automatically. The three spatial derivatives of the signal phase surrounding a field‐perturbing object allowed three‐dimensional localization, as well as the discrimination of diamagnetic field‐perturbing objects, such as calcifications, and paramagnetic field‐perturbing structures, e.g. blood. A longer echo time enabled smaller disturbances to be detected, but also resulted in shading as a result of other field‐disturbing materials. A higher imaging resolution increased the detection sensitivity. Microcalcifications in a linear branching configuration that spanned over 8 mm in length were detected. After manual correction, the automatic detection tool identified up to 18 microcalcifications within the samples, which was in close agreement with the number of microcalcifications found on previously acquired in vivo mammography images. Microcalcifications can be detected by MRI in human whole breast specimens by the application of phase derivative imaging. Copyright © 2014 John Wiley & Sons, Ltd.