This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Computer applications, Oncology, MR-Diffusion/Perfusion, Computer Applications-Detection, diagnosis, Neoplasia, Abscess, Metastases
Authors:
G. C. Manikis1, K. Nikiforaki1, G. Ioannidis1, N. Papanikolaou2, K. Marias1; 1Heraklion/GR, 2Lisbon/PT
DOI:
10.1594/ecr2017/C-2964
Aims and objectives
This study aims to elucidate aspects of IVIM fitting procedure that may obscure the true contribution of pseudo-diffusion to a multi-b-value DWI acquisition scheme. The vast majority concludes on the significance of D and f in aiding diagnosis,
while the evaluated D* values are rarely used as basic criterion for differentiation between clinical entities [1].
Specifically,
the fitted value for D* tends to coincide with its upper boundary value in a standard non-linear least squares (NLLS) fitting method,
being the most sensitive to fitting errors among the IVIM parameters.
The scope of this work is mainly relied on a three-fold rationale; a) to propose an optimized method for fitting the DW-MRI data according to the IVIM model in order to reduce pixel percentage converging to the upper bound value of D*,
b) to statistically assess the goodness-of-fit of the proposed method and compare the results with commonly fitting methods used under a variety of SNR levels,
and c) to estimate the ability of the optimized fitting version in providing repeatable D* measurements with low variability.