Fulltext

Quantitative breast MRI: 2D histogram analysis of diffusion tensor parameters in normal tissue

Wiederer, Julia ; Pazahr, Shila ; Leo, Cornelia ; Nanz, Daniel ; Boss, Andreas

In: Magnetic Resonance Materials in Physics, Biology and Medicine, 2014, vol. 27, no. 2, p. 185-193

Add to personal list
    Title
    • Quantitative breast MRI: 2D histogram analysis of diffusion tensor parameters in normal tissue
    Author
    • Wiederer, Julia. Institute of Diagnostic and Interventional Radiology, University Hospital of Zurich, Raemistr. 100, 8091, Zurich, Switzerland
    • Pazahr, Shila. Institute of Diagnostic and Interventional Radiology, University Hospital of Zurich, Raemistr. 100, 8091, Zurich, Switzerland
    • Leo, Cornelia. Division of Gynecology, University Hospital of Zurich, Zurich, Switzerland
    • Nanz, Daniel. Institute of Diagnostic and Interventional Radiology, University Hospital of Zurich, Raemistr. 100, 8091, Zurich, Switzerland
    • Boss, Andreas. Institute of Diagnostic and Interventional Radiology, University Hospital of Zurich, Raemistr. 100, 8091, Zurich, Switzerland
    Document Type
    • Postprint
    Language
    • English
    Published in
    • Magnetic Resonance Materials in Physics, Biology and Medicine, 2014, vol. 27, no. 2, p. 185-193. Springer Berlin Heidelberg
    Other Version
    OAI-PMH Identifier
    • oai:doc.rero.ch:325293
    Summary
    Abstract : Object: Diffusion tensor imaging (DTI) of the breast may provide a powerful new approach for the detection of intraductal processes. The aim of this investigation was to characterize the relation between diffusion tensor parameters [fractional anisotropy (FA), mean diffusivity (MD)] in normal breast tissue to obtain information on the microenvironment of the diffusing water molecules and to provide a systematic approach for DTI analysis. Materials and methods: Seven female, healthy volunteers underwent prospective double-spin-echo prepared echo-planar diffusion-weighted sequence (TR/TE 8,250ms/74ms, b values 0 and 500s/mm (2), six encoding directions, 12 averages, 35 slices) in 4 consecutive weeks (3.0 T). Quantitative maps of diffusion tensor parameters were computed offline with custom routines. The interdependence of MD and FA in different voxels was analysed by linear and exponential regression. Results: All MD and FA maps were of excellent quality. A consistent pattern was observed in that lower fractional anisotropy values were more likely associated with higher mean diffusivity values. The dependence exhibited an exponential behavior with a correlation coefficient R=0.60 (R linear=0.57). Conclusion: The likelihood with which FA and MD values are observed in a voxel within normal breast tissue is characterized by a specific pattern, which can be described by an exponential model. Moreover, we could show that the proposed technique does not depend on the menstrual cycle.