Herve Saint-Jalmes PhD, Presenter: Nothing to Disclose

Benjamin Leporq MS, Abstract Co-Author: Nothing to Disclose

Frank Pilleul MD, Abstract Co-Author: Nothing to Disclose

Olivier Beuf PhD, Abstract Co-Author: Nothing to Disclose

To optimize a 3.0T acquisition protocol for liver Intra-Voxel Incoherent Motion imaging (IVIM) imaging to be included in a clinical study focused on chronic liver diseases.

First, acquisition protocol was evaluated on 25 healthy volunteers (16 men, 9 women; mean age: 27.1 years; mean weight: 71 kg). Acquisitions were performed on a 3.0T GE Discovery MR 750 (GEHC, Milwaukee, WI, USA) system with 50 mT/m maximum gradient amplitude. Sequence used was the SE-EPI eDWI sequence (enhanced Diffusion Weighted Imaging) including 12 b-factors (0, 10, 20, 40, 60, 80, 100, 200, 300, 400, 600, and 800 s/mm2) with variable NEX according to b-factor (2-2-2-2-3-3-4-5-6-7-8-9 NEX) with 2000 ms TR and 55 ms TE. 21 axial slices were acquired with a 400 × 300 mm² FOV, 128 × 96 acquisition matrix, 8 mm slice thickness, and a 250 KHz bandwidth. Fat Sat was disabled. Signal was collected using the 32 channels body coil. Then scan duration was 5’12”. Second, another optimization was lead to minimize the number of b-values required to keep the same data fit reliability from the protocol using Cramér–Rao inequality providing optimized b-factors. Pure molecular diffusion coefficient (Dslow), perfusion-related coefficient (Dfast) and perfusion fraction (f) were obtained by a non-linear least-square fit to the bi-exponential IVIM model.

Theoritical optimization based on Cramér–Rao suggested the used of 0, 10, 80 and 800 s/mm2 b-values. Mean IVIM parameters obtained were: Dslow value of 1.08 (10-3 mm2/s), 21% for f and 79.3 to 87.9 (10-3 mm2/s) for Dfast. Bland-Altman plots showed no significant differences between 12b- and 4b-methods.

Minimization study showed that using a limited (4) number of chosen b-factors give the same results compared with 12 b-factors. The use of SmartNex, 3-in-1 diffusion gradient scheme, free breathing technique and only 4 b-factors enabled whole liver Intra Voxel Incoherent Motion Imaging within a minute and seem to be a suitable compromise to be added in a clinical protocol focusing on chronic liver diseases assessment.

Optimized MRI protocol at 3.0T dedicated to liver examination of chronic liver diseases focusing on fibrosis and cirrhosis.

Saint-Jalmes, H, Leporq, B, Pilleul, F, Beuf, O, Chronic Liver Diseases Assessment with Optimized Intra-Voxel Incoherent Motion MRI Protocol at 3.0T.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13017908.html