Premium
Assessment of Myocardial Architecture Depends on Imaging DTMRI Parameters
Author(s) -
Agger Peter,
Laustsen Christoffer,
Hjortdal Vibeke,
Pedersen Michael
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.780.5
Subject(s) - fractional anisotropy , diffusion mri , anisotropy , voxel , nuclear magnetic resonance , magnetic resonance imaging , biomedical engineering , mathematics , nuclear medicine , medicine , physics , optics , radiology
Background Diffusion tensor magnetic resonance imaging (DTMRI) is widely used when assessing the orientation of cardiomyocytes in the heart. Three important parameters are obtained from DTMRI in order to define the myocardial anatomical substructures: fractional anisotropy (FA), mean diffusivity (MD), and helical angle (HA). The aim of this study was to demonstrate the effect of DTMRI sequence parameters: averaging, anisotropic resolution and number of diffusion‐weighted directions, on FA, MD and HA. Methods Health adult porcine hearts were harvested, formalin‐fixed, and a transmural biopsy of the left ventricular free wall was sampled and positioned in a 9.4 T MRI system. Five different DTMRI examinations were conducted: 1) with 6 predefined gradient directions, 2) with 30 directions, 3) with 6 directions and 2 averages, 4+5) with 6 and 30 directions, but with double slice thickness rendering the voxels anisotropic. All other DTMRI imaging parameters were kept constant. Differences in FA, MD and HA were analysed using ANOVA statistics. Results MD showed only minor non‐significant differences between groups. FA was decreased when using 30 diffusion directions (0.25±0.08) as compared with 6 directions (0.33±0.13), p<0.001. Helical angulations as compared using Bland‐Altman plots ( Fig. 1) confirmed this finding revealing a variance indicative of low precision when using only 6 directions. Tractographies based on only 6 directions were less smooth and shorter when compared with 30 directions ( Fig. 2). Conclusions This DTMRI study revealed that fewer gradient directions resulted in deterioration of the image quality, increasing the degree of averaging does not compensate for this. Introduction of anisotropic voxels on the other hand did not alter FA, MD or HA significantly. The results of this study are of importance when designing DTMRI protocols for the purpose of investigating myocardial microanatomy. We emphasize that interpretation of DTMRI results depends on the number of predefined diffusion directions. Support or Funding Information This work is supported by the Danish Children Heart Foundation.