On the Least-Square Estimation of Parameters for Statistical Diffusion Weighted Imaging Model
Refereed conference paper presented and published in conference proceedings

Full Text

Times Cited
Web of Science2WOS source URL (as at 23/11/2020) Click here for the latest count

Other information
AbstractStatistical model for diffusion-weighted imaging (DWI) has been proposed for better tissue characterization by introducing a distribution function for apparent diffusion coefficients (ADC) to account for the restrictions and hindrances to water diffusion in biological tissues. This paper studies the precision and uncertainty in the estimation of parameters for statistical DWI model with Gaussian distribution, i. e. the position of distribution maxima (D-m) and the distribution width (sigma), by using non-linear least-square (NLLS) fitting. Numerical simulation shows that precise parameter estimation, particularly for sigma, imposes critical requirements on the extremely high signal-to-noise ratio (SNR) of DWI signal when NLLS fitting is used. Unfortunately, such extremely high SNR may be difficult to achieve for the normal setting of clinical DWI scan. For D-m and sigma parameter mapping of in vivo human brain, multiple local minima are found and result in large uncertainties in the estimation of distribution width sigma. The estimation error by using NLLS fitting originates primarily from the insensitivity of DWI signal intensity to distribution width sigma, as given in the function form of the Gaussian-type statistical DWI model.
All Author(s) ListYuan J, Zhang QW
Name of Conference35th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC)
Start Date of Conference03/07/2013
End Date of Conference07/07/2013
Place of ConferenceOsaka
Country/Region of ConferenceJapan
Journal nameConference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Detailed descriptionorganized by IEEE, EMB, JSMBE,
Pages4406 - 4409
LanguagesEnglish-United Kingdom
Web of Science Subject CategoriesEngineering; Engineering, Biomedical; Engineering, Electrical & Electronic

Last updated on 2020-24-11 at 00:38