Table of Contents
About the Editors xi
List of Contributors xiii
1 Introduction 1
2 Evolutionary Computation: A Brief Overview Stefano Cagnoni Leonardo Vanneschi 3
2.1 Introduction 3
2.2 Evolutionary Computation Paradigms 4
2.2.1 Genetic Algorithms 5
2.2.2 Evolution Strategies 7
2.2.3 Evolutionary Programming 8
2.2.4 Genetic Programming 8
2.2.5 Other Evolutionary Techniques 10
2.2.6 Theory of Evolutionary Algorithms 11
2.3 Conclusions 12
3 A Review of Medical Applications of Genetic and Evolutionary Computation Stephen L. Smith 17
3.1 Medical Imaging and Signal Processing 18
3.1.1 Overview 18
3.1.2 Image Segmentation 18
3.1.3 Image Registration, Reconstruction and Correction 21
3.1.4 Other Applications 24
3.2 Data Mining Medical Data and Patient Records 25
3.3 Clinical Expert Systems and Knowledge-based Systems 27
3.4 Modelling and Simulation of Medical Processes 29
3.5 Clinical Diagnosis and Therapy 34
4 Applications of GEC in Medical Imaging 45
4.1 Evolutionary Deformable Models for Medical Image Segmentation: A Genetic Algorithm Approach to Optimizing Learned, Intuitive, and Localized Medial-based Shape Deformation Chris McIntosh Ghassan Hamarneh 47
4.1.1 Introduction 47
4.1.1.1 Statistically Constrained Localized and Intuitive Deformations 54
4.1.1.2 Genetic Algorithms 57
4.1.2 Methods 58
4.1.2.1 Population Representation 58
4.1.2.2 Encoding the Weights for GAs 58
4.1.2.3 Mutations and Crossovers 59
4.1.2.4 Calculating the Fitness of Members of the GA Population 60
4.1.3 Results 62
4.1.4 Conclusions 63
4.2 Feature Selection for the Classification of Microcalcifications in Digital Mammograms using Genetic Algorithms, Sequential Search and Class Separability Santiago E. Conant-Pablos Rolando R. Hernández-Cisneros Hugo Terashima-Marín 69
4.2.1 Introduction 69
4.2.2 Methodology 71
4.2.2.1 Pre-processing 71
4.2.2.2 Detection of Potential Microcalcifications (Signals) 72
4.2.2.3 Classification of Signals into Microcalcifications 74
4.2.2.4 Detection of Microcalcification Clusters 76
4.2.2.5 Classification of Microcalcification Clusters into Benign and Malignant 77
4.2.3 Experiments and Results 77
4.2.3.1 From Pre-processing to Signal Extraction 77
4.2.3.2 Classification of Signals into Microcalcifications 79
4.2.3.3 Microcalcification Clusters Detection and Classification 81
4.2.4 Conclusions and Future Work 82
4.3 Hybrid Detection of Features within the Retinal Fundus using a Genetic Algorithm Vitoantonio Bevilacqua Lucia Cariello Simona Cambò Domenico Daleno Giuseppe Mastronardi 85
4.3.1 Introduction 85
4.3.2 Acquisition and Processing of Retinal Fundus Images 88
4.3.2.1 Retinal Image Acquisition 89
4.3.2.2 Image Processing 90
4.3.3 Previous Work 91
4.3.4 Implementation 93
4.3.4.1 Vasculature Extraction 94
4.3.4.2 A Genetic Algorithm for Edge Extraction 99
4.3.4.3 Skeletonization Process 103
4.3.4.4 Experimental Results 104
5 New Analysis of Medical Data Sets using GEC 111
5.1 Analysis and Classification of Mammography Reports using Maximum Variation Sampling Robert M. Patton Barbara G. Beckerman Thomas E. Potok 113
5.1.1 Introduction 113
5.1.2 Background 114
5.1.3 Related Works 116
5.1.4 Maximum Variation Sampling 118
5.1.5 Data 122
5.1.6 Tests 124
5.1.7 Results & Discussion 124
5.1.8 Summary 129
5.2 An Interactive Search for Rules in Medical Data using Multiobjective Evolutionary Algorithms Daniela Zaharie D. Lungeanu Flavia Zamfirache 133
5.2.1 Medical Data Mining 133
5.2.2 Measures for Evaluating the Rules Quality 134
5.2.2.1 Accuracy Measures 135
5.2.2.2 Comprehensibility Measures 135
5.2.2.3 Interestingness Measures 136
5.2.3 Evolutionary Approaches in Rules Mining 137
5.2.4 An Interactive Multiobjective Evolutionary Algorithm for Rules Mining 138
5.2.4.1 Rules Encoding 139
5.2.4.2 Reproduction Operators 139
5.2.4.3 Selection and Archiving 140
5.2.4.4 User Guided Evolutionary Search 141
5.2.5 Experiments in Medical Rules Mining 143
5.2.5.1 Impact of User Interaction 144
5.2.6 Conclusions 146
5.3 Genetic Programming for Exploring Medical Data using Visual Spaces Julio J. Valdés Alan J. Barton Robert Orchard 149
5.3.1 Introduction 149
5.3.2 Visual Spaces 150
5.3.2.1 Visual Space Realization 150
5.3.2.2 Visual Space Taxonomy 150
5.3.2.3 Visual Space Geometries 151
5.3.2.4 Visual Space Interpretation Taxonomy 151
5.3.2.5 Visual Space Characteristics Examination 153
5.3.2.6 Visual Space Mapping Taxonomy 154
5.3.2.7 Visual Space Mapping Computation 155
5.3.3 Experimental Settings 157
5.3.3.1 Implicit Classical Algorithm Settings 158
5.3.3.2 Explicit GEP Algorithm Settings 159
5.3.4 Medical Examples 161
5.3.4.1 Data Space Examples 161
5.3.4.2 Semantic Space Examples 164
5.3.5 Future Directions 170
6 Advanced Modelling, Diagnosis and Treatment using GEC 173
6.1 Objective Assessment of Visuo-spatial Ability using Implicit Context Representation Cartesian Genetic Programming Michael A. Lones Stephen L. Smith 175
6.1.1 Introduction 175
6.1.2 Evaluation of Visuo-spatial Ability 176
6.1.3 Implicit Context Representation CGP 178
6.1.4 Methodology 180
6.1.4.1 Data Collection 181
6.1.4.2 Evaluation 181
6.1.4.3 Parameter Settings 182
6.1.5 Results 184
6.1.6 Conclusions 186
6.2 Towards an Alternative to Magnetic Resonance Imaging for Vocal Tract Shape Measurement using the Principles of Evolution David M. Howard Andy M. Tyrrell Crispin Cooper 191
6.2.1 Introduction 191
6.2.2 Oral Tract Shape Evolution 194
6.2.3 Recording the Target Vowels 195
6.2.4 Evolving Oral Tract Shapes 196
6.2.5 Results 199
6.2.5.1 Oral Tract Areas 200
6.2.5.2 Spectral Comparisons 201
6.2.6 Conclusions 204
6.3 How Genetic Algorithms can Improve Pacemaker Efficiency Laurent Dumas Linda El Alaoui 209
6.3.1 Introduction 209
6.3.2 Modeling of the Electrical Activity of the Heart 211
6.3.3 The Optimization Principles 213
6.3.3.1 The Cost Function 213
6.3.3.2 The Optimization Algorithm 213
6.3.3.3 A New Genetic Algorithm with a Surrogate Model 214
6.3.3.4 Results of AGA on Test Functions 215
6.3.4 A Simplified Test Case for a Pacemaker Optimization 216
6.3.4.1 Description of the Test Case 216
6.3.4.2 Numerical Results 218
6.3.5 Conclusion 220
7 The Future for Genetic and Evolutionary Computation in Medicine: Opportunities, Challenges and Rewards 223
7.1 Opportunities 224
7.2 Challenges 224
7.3 Rewards 226
7.4 The Future for Genetic and Evolutionary Computation in Medicine 227
Appendix: Introductory Books and Useful Links 229
Index 231