Advanced Medical Biometrics eBook

In brief, this book provides the advanced sensing technologies including multi-sensor collaborative sensing for pulse signal acquisition, sensor array design for breath odor perception, and voice signal acquisition. Besides, this book summarizes the advances about pulse feature extraction with improved Gabor function, odor feature extraction based on de-convolution methods. This book also presents the latest recognition methods including evolutionary ensemble model to handle the imbalance problem, multi-feature complementary methods for face image, tongue images, and pulse analysis. Multi-modal hybrid fusion and self-adaptive score fusion for disease classification are proposed and proved effective for medical biometrics.

Medical biometrics achieves disease detection and health monitor by sensing and recognizing multi-modal body surface information. It is believed that body surface information, e.g., face, tongue, breath odor, voice, and wrist pulse, is one of the most important biometric features that contains rich pathological information and can be used for health assessments or disease detection. Focusing on the latest progresses in medical biometrics, this book systematically presents advanced signal acquisition techniques, feature extraction methods, and recognition algorithms.

Both researchers and engineers in pattern recognition and medical diagnosis will benefit from this book as it offers comprehensive understanding of the advanced signal acquisition techniques, related feature extraction methods, and state-of-the-art analysis methods.

Contents:

• Preface
• About the Authors
• Introduction to Advanced Medical Biometrics
Tongue and Face Image Analysis:
• Evolutionary Ensemble Model for Disease Detection Using Tongue Images
• Color Representation and Feature Fusion for Diagnosis Using Sublingual Images
• Multi-feature Learning Model for Diagnosis Using Face Skin Images
Pulse Signal Analysis:
• A Robust Wrist Pulse Acquisition System Based on Multi-sensor Collaboration and Signal Quality Assessment
• Pressure Wrist Pulse Signal Analysis by Sparse Decomposition Using Improved Gabor Function
• Multi-feature Fusion Model for Diabetes Mellitus Detection Using Pulse Signals
Odor Analysis:
• A Data-driven Study on Pathological Explanation for Odor-based Disease Detection
• A Novel Medical E-Nose Signal Analysis System
• LTI-ODeC: A Novel De-convolution Feature Extraction Method for Odor-based Multi-disease Diagnosis
Voice Analysis:
• Voice Signal Acquisition System for Disease Detection
Multi-modal Fusion for Diagnosis:
• Multi-modal Hybrid Fusion Analysis on Heterogeneous Non-invasive Traditional Chinese Medicine Diagnosis
• Self-adaptive Score Fusion for Multi-modal Disease Classification
• Book Review and Future Work
• Bibliography
• Index

Readership: Graduate and senior undergraduate students from computer science, artificial intelligent, and machine learning, or those interested in pattern recognition and medical biometrics; Medical researchers and students.