Abstract
The package dcemriS4 provides a complete set of data analysis tools for quantitative assessment of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Image processing is provided for the ANALYZE and NIfTI data formats as input with all parameter estimates being output in NIfTI format. Estimation of T1 relaxation from multiple flip-angle acquisitions, using either constant or spatially-varying flip angles, is performed via nonlinear regression. Both literature-based and data-driven arterial input functions are available and may be combined with a variety of compartmental models. Kinetic parameters are obtained from nonlinear regression, Bayesian estimation via Markov chain Monte Carlo or Bayesian maximum a posteriori estimation. A non-parametric model, using penalized splines, is also available to characterize the contrast agent concentration time curves. Estimation of the apparent diffusion coefficient (ADC) is provided for diffusion-weighted imaging. Given the size of multi-dimensional data sets commonly acquired in imaging studies, care has been taken to maximize computational efficiency and minimize memory usage. All methods are illustrated using both simulated and real-world medical imaging data available in the public domain.
Highlights
Quantitative analysis of tissue perfusion using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is achieved through a series of processing steps, starting with the raw data acquired from the MRI scanner, and involves a combination of physics, mathematics, engineering and statistics to produce a set of statistical images based on parameter estimates from a compartmental model
Quantitative analysis of dynamic contrast-enhanced MRI (DCE-MRI) and diffusion-weighted imaging (DWI) data requires a series of processing steps, including pre-processing of the MR signal, voxel-wise curve fitting, and post-processing
The dcemriS4 package provides a comprehensive set of functions for pre-processing and parametric models for quantifying DCE-MRI and Diffusion weighted imaging (DWI) data
Summary
Quantitative analysis of tissue perfusion using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is achieved through a series of processing steps, starting with the raw data acquired from the MRI scanner, and involves a combination of physics, mathematics, engineering and statistics to produce a set of statistical images based on parameter estimates from a compartmental model. The S4 designation in dcemriS4 means that S4 object classes are used throughout to ensure efficient and transparent manipulation of ANALYZE or NIfTI data structures. Parameter estimates in dcemriS4 inherit attributes from the incoming ANALYZE/NIfTI objects in order to preserve anatomical and physiological information for appropriate visualization. All functions for parameter estimation may be applied to aggregated data; i.e., a mean curve across an anatomical region of interest. As voxel-wise quantitative analysis can be time consuming, dcemriS4 supports basic parallel computing by incorporating the multicore package (Urbanek 2011).
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call