Meeting Banner
Abstract #4039

A Systematic Technique for Echo Truncation in MRI-Based R2* Calculations for Liver Iron Quantification Which Reduces Systematic and Random Error

Mustafa Rifaat Bashir1, Brian M. Dale2, Elmar Max Merkle1

1Radiology, Duke University Medical Center, Durham, NC, United States; 2Siemens Medical Solutions, Cary, NC, United States

Multi-echo gradient echo based MRI techniques have emerged as a useful clinical tool for estimating hepatic iron concentration by measuring liver R2*. Systematic and random errors can occur in the R2* solution based on the number of echoes used for the calculation. We explore the propagation of these errors, and propose and analyze a technique for reducing error by selecting an optimized number of echoes for the calculation. We compare this with standard techniques using Monte Carlo simulations and patient data.

Abstract PDF E-Poster Video

Keywords

according acquisition applying assumes basis best bias blood board calculation cause choice chosen clinical considered custom cutoff dale database decays described deviation diffusion distribution divided duke echoes emerged employed encountered equation equations error errors estimations exam fact fitting five fixed floor fluctuating gradient greater health institutional intensity involve iron know known larger lead least linear liver local location long lowest magnitude maps mate mated measured medical meed mine minimize noise outperform pair paradigm patient phenomenon positive predetermined priori problem progressive propose random reduced reduces refer resultant review selected sets simulated simulation simulations since solution spacing squares symmetric system systematic tech tended took trends undergone variable variance whole wood