Statistical Analysis of Noise in MRI: Modeling, Filtering and Estimation

ISBN-13: 9783319820002 / Angielski / Miękka / 2018 / 327 str.

Santiago Aja-Fernandez; Gonzalo Vegas-Sánchez-Ferrero

Opis

This unique text presents a comprehensive review of methods for modeling signal and noise in magnetic resonance imaging (MRI), providing a systematic study, classifying and comparing the numerous and varied estimation and filtering techniques. Features: provides a complete framework for the modeling and analysis of noise in MRI, considering different modalities and acquisition techniques; describes noise and signal estimation for MRI from a statistical signal processing perspective; surveys the different methods to remove noise in MRI acquisitions from a practical point of view; reviews different techniques for estimating noise from MRI data in single- and multiple-coil systems for fully sampled acquisitions; examines the issue of noise estimation when accelerated acquisitions are considered, and parallel imaging methods are used to reconstruct the signal; includes appendices covering probability density functions, combinations of random variables used to derive estimators, and useful MRI datasets.

Recenzje

"The book is presented in a simple and lucid manner, starting with the basics of MRI noise and its analysis with simple models, progressing to an analysis using complex models and the noise issues in multi-coil and parallel acquisition schemes. Overall the book is self-contained to help the beginners ... ." (Pramod Kumar Pisharady, IAPR Newsletter , Vol. 40 (2), 2018)

Spis treści

The Problem of Noise in MRI

Part I: Noise Models and the Noise Analysis Problem

Acquisition and Reconstruction of Magnetic Resonance Imaging

Statistical Noise Models for MRI

Noise Analysis in MRI: Overview

Noise Filtering in MRI

Part II: Noise Analysis in Non-Accelerated Acquisitions

Noise Estimation in the Complex Domain

Noise Estimation in Single-Coil MR Data

Noise Estimation in Multiple-Coil MR Data

Parametric Noise Analysis from Correlated Multiple-Coil MR Data

Part III: Noise Estimators in pMRI

Parametric Noise Analysis in Parallel MRI

Blind Estimation of Non-Stationary Noise in MRI

Appendix A: Probability Distributions and Combination of Random Variables

Appendix B: Variance Stabilizing Transformation

Appendix C: Data Sets Used in the Experiments

Blurb

This unique text/reference presents a comprehensive review of methods for modeling signal and noise in magnetic resonance imaging (MRI), providing a systematic study, classifying and comparing the numerous and varied estimation and filtering techniques drawn from more than ten years of research in this area.

Topics and features:

• Provides a complete framework for the modeling and analysis of noise in MRI, considering different modalities and acquisition techniques
• Describes noise and signal estimation for MRI from a statistical signal processing perspective
• Surveys the different methods to remove noise in MRI acquisitions, under different approaches and from a practical point of view
• Reviews different techniques for estimating noise from MRI data in single- and multiple-coil systems for fully sampled acquisitions
• Examines the issue of noise estimation when accelerated acquisitions are considered, and parallel imaging methods are used to reconstruct the signal
• Includes appendices covering probability density functions, combinations of random variables used to derive estimators, and useful MRI datasets

This practically-focused work serves as a reference manual for researchers dealing with signal processing in MRI acquisitions, and is also suitable as a textbook for postgraduate students in engineering with an interest in medical image processing.

Dr. Santiago Aja-Fernández is an Associate Professor at the School of Telecommunications of the University of Valladolid, Spain. His other publications include the Springer title Tensors in Image Processing and Computer Vision. Dr. Gonzalo Vegas-Sánchez-Ferrero is a Research Fellow at Brigham and Women’s Hospital, and in the Applied Chest Imaging Laboratory of Harvard Medical School, Boston, MA, USA.