High Performance Deformable Image Registration Algorithms for Manycore Processors develops highly data-parallel image registration algorithms suitable for use on modern multi-core architectures, including graphics processing units (GPUs). Focusing on deformable registration, we show how to develop data-parallel versions of the registration algorithm suitable for execution on the GPU. Image registration is the process of aligning two or more images into a common coordinate frame and is a fundamental step to be able to compare or fuse data obtained from different sensor measurements. Extracting useful information from 2D/3D data is essential to realizing key technologies underlying our daily lives. Examples include autonomous vehicles and humanoid robots that can recognize and manipulate objects in cluttered environments using stereo vision and laser sensing and medical imaging to localize and diagnose tumors in internal organs using data captured by CT/MRI scans.
Key Features
Demonstrates how to redesign widely used image registration algorithms so as to best expose the underlying parallelism available in these algorithms
Shows how to pose and implement the parallel versions of the algorithms within the single instruction, multiple data (SIMD) model supported by GPUs
Provides Programming "tricks" that can help readers develop other image processing algorithms, including registration algorithms for the GPU

Biographies vii

Chapter 1 Introduction 1

1.1 Introduction 1

1.2 Applications of Deformable Image Registration 4

1.3 Algorithmic Approaches to Deformable Registration 5

1.4 Organization of Chapters 7

References 9

Chapter 2 Unimodal B-Spline Registration 13

2.1 Introduction 13

2.2 Overview of B-Spline Registration 16

2.3 B-Spline Registration on the GPU 25

2.4 Performance Evaluation 35

2.5 Summary 40

References 43

Chapter 3 Multimodal B-Spline Registration 45

3.1 Introduction 45

3.2 Using B-Splines to Represent the Deformation Field 47

3.3 MI as A Cost Function 51

3.4 Efficient Computation of MI 54

3.5 Performance Evaluation 68

3.6 Related Work 72

3.7 Summary 73

References 74

Chapter 4 Analytic Vector Field Regularization for B-spline Parameterized Methods 75

4.1 Introduction 75

4.2 Theory and Mathematical Formalism 75

4.3 Algorithmic Implementation 80

4.4 Performance Evaluation 84

4.5 Summary 91

References 93

Chapter 5 Deformable Registration Using Optical-Flow Methods 95

5.1 Introduction 95

5.2 Demons Algorithm for Deformable Registration 96

5.3 SIMD Version of Demons Algorithm 97

5.4 Performance Evaluation 102

5.5 Summary 106

References 106

Chapter 6 Plastimatch—An Open-Source Software for Radiotherapy Imaging 107

6.1 Introduction 107

6.2 Overview of Plastimatch 108

6.3 Licensing 113

References 113

Greg Sharp is a computer scientist and medical physicist at Massachusetts General Hospital. He received his Ph.D. in the depart-ment of electrical engineering and computer science from the University of Michigan in 2002, and currently holds an appointment of assistant professor of radiation oncology at Harvard Medical School. Prof. Sharp's research interests include medical image computing, image-guided radiation therapy, and motion management.

中科院文献情报中心