●Chapter 1 Introduction 1
1.1 Breast Cancer Status 1
1.2 Mammography 2
1.3 Mammographic Risk Assessment 4
1.3.1 Wolfe’s Four Risk Categories 4
1.3.2 Boyd’s Six Class Categories 5
1.3.3 Four BIRADS Density Categories 5
1.3.4 Tabár’s Five Patterns 5
1.4 CAD in Mammography 7
1.5 Clinical Utility of the Present Research 8
1.6 Focus and Contributions of the Book 8
1.7 Book Outline 10
Chapter 2 A Literature Review of Mammographic Image Analysis 12
2.1 Mammographic Image Segmentation 12
2.1.1 Breast Region Segmentation 12
2.1.2 Breast Density Segmentation 19
2.2 Estimation of Mammographic Density 23
2.3 Characterisation of Mammographic Parenchymal Patterns 28
2.4 Breast Density Classification 33
2.5 Summary 37
Chapter 3 Image Segmentation in Mammography 38
3.1 Breast Region Segmentation in Mammograms 38
3.1.1 Methodology 38
3.1.2 Results and Discussion 42
3.2 A Modified FCM Algorithm for Breast Density Segmentation 49
3.2.1 FCM Algorithms 49
3.2.2 A Modified FCM Algorithm 51
3.2.3 Experimental Results 53
3.3 Topographic Representation Based Breast Density Segmentation 57
3.3.1 Topographic Representation 57
3.3.2 Segmentation of Dense Tissue Regions 59
3.3.3 Breast Density Quantification 61
3.3.4 Results 62
3.4 Summary 64
Chapter 4 Texture Analysis in Mammography 66
4.1 Local Feature Based Texture Representations 66
4.1.1 Local Binary Patterns 67
4.1.2 Local Grey-Level Appearances 67
4.1.3 Basic Image Features 68
4.1.4 Textons 69
4.2 Mammographic Tissue Appearance Modelling 70
4.3 Summary 74
Chapter 5 Multiscale Blob Detection in Mammography 75
5.1 Blob Detection 75
5.1.1 Laplacian of Gaussian 75
5.1.2 Difference of Gaussian 76
5.1.3 Determinant of the Hessian Matrix 76
5.1.4 Hessian-Laplacian 77
5.1.5 Fast-Hessian 77
5.1.6 Salient Region 77
5.2 A Blob Based Representation of Mammographic Parenchymal Patterns 78
5.2.1 Detection of Multiscale Blobs 79
5.2.2 Blob Merging 85
5.2.3 Blob Encoding 88
5.3 Results and Discussion 88
5.4 Summary 93
Chapter 6 Breast Cancer Risk Assessment 95
6.1 Experimental Data 95
6.1.1 MIAS Database 95
6.1.2 DDSM Database 96
6.2 Evaluation Methodology 97
6.2.1 Classification Algorithm 97
6.2.2 Cross-Validation Scheme 98
6.2.3 Result Representation 100
6.3 Evaluating the Proposed Methods 100
6.3.1 Evaluation of Breast Density Segmentation 100
6.3.2 Evaluation of Breast Tissue Appearance Modelling 108
6.3.3 A Combined Modelling of Breast Tissue 112
6.3.4 Evaluation of Blob-Based Representation 115
6.4 Summary 118
Chapter 7 Discussions on Breast Cancer Risk Assessment in Mammography 120
7.1 Comparison of the Proposed Methods 120
7.2 Comparing with Related Publications 126
7.3 Summary 130
Chapter 8 Computer-Aided Diagnosis of Breast Cancer Based on Deep Learning 131
8.1 Literature Review on Deep Learning Based Mammographic Image Analysis 131
8.2 Mass Detection and Classification in Mammograms withaDeepPipeline 135
8.2.1 Dataset Information 136
8.2.2 Model Architecture 139
8.2.3 Training 140
8.2.4 Results & Discussion 140
8.3 Summary 149
Chapter 9 Conclusions 150
9.1 Summary of the Book 150
9.2 Contributions and Novel Aspects 152
9.3 Future Work 154
Bibliography 156
Biography 167
内容简介
Breast cancer is currently the most common cancer among women worldwide. Mammography is the most reliable and effective tool for the screening and early detection of breast cancer. Recently, with the rapid development of artifi intelligence technology, computer-aided diagnosis technology has become the main research topic in the field of artifi intelligence plus medical imaging. In this book, the ......
