Human Motion Capture and Identification for Assistive Systems Design in Rehabilitation

Cover
Title Page
Copyright
Contents
Chapter 1 Introduction
	1.1 Human Body – Kinematic Perspective
	1.2 Musculoskeletal Injuries and Neurological Movement Disorders
		1.2.1 Musculoskeletal injuries
		1.2.2 Neuromuscular disorders
	1.3 Sensors in Telerehabilitation
		1.3.1 Opto‐electronic sensing
		1.3.2 RGB camera and microphone
		1.3.3 Inertial measurement unit (IMU)
	1.4 Model‐based State Estimation and Sensor Fusion
		1.4.1 Summary and challenges
	1.5 Human Motion Encoding in Telerehabilitation
		1.5.1 Human motion encoders in action recognition
		1.5.2 Human motion encoders in physical telerehabilitation
		1.5.3 Summary and challenge
	1.6 Patients' Performance Evaluation
		1.6.1 Questionnaire‐based assessment scales
		1.6.2 Automated kinematic performance assessment
		1.6.3 Summary and challenge
Chapter 2 Kinematic Performance Evaluation with Non‐wearable Sensors
	2.1 Introduction
	2.2 Fusion
		2.2.1 Introduction
		2.2.2 Linear model of human motion multi‐Kinect system
		2.2.3 Model‐based state estimation
		2.2.4 Fusion of information
		2.2.5 Mitigation of occlusions and optimised positioning
		2.2.6 Computer simulations and hardware implementation
	2.3 Encoder
		2.3.1 Introduction
		2.3.2 The two‐component encoder theory
		2.3.3 Encoding methods
		2.3.4 Dealing with noise
		2.3.5 Complex motion decomposition using switching continuous hidden Markov models
		2.3.6 Canonical actions and the action alphabet
		2.3.7 Experiments and results
	2.4 ADL Kinematic Performance Evaluation
		2.4.1 Introduction
		2.4.2 Methodology
		2.4.3 Experiment setup
		2.4.4 Data analysis and results
	2.5 Summary
Chapter 3 Biokinematic Measurement with Wearable Sensors
	3.1 Introduction
	3.2 Introduction to Quaternions
	3.3 Wahba's Problem
		3.3.1 Solutions to the Wahba problem
		3.3.2 Davenport's q method
		3.3.3 Quaternion Estimation Algorithm (QUEST)
		3.3.4 Fast optimal attitude matrix (FOAM)
		3.3.5 Estimator of the optimal quarternion (ESOQ or ESOQ1) method
	3.4 Quaternion Propagation
	3.5 MARG (Magnetic Angular Rates and Gravity) Sensor Arrays‐based Algorithm
	3.6 Model‐based Estimation of Attitude with IMU Data
	3.7 Robust Optimisation‐based Approach for Orientation Estimation
	3.8 Implementation of the Orientation Estimation
		3.8.1 Extended Kalman filter‐based approach
		3.8.2 Robust extended Kalman filter implementation
		3.8.3 Robust extended Kalman filter with linear measurements
	3.9 Computer Simulations
	3.10 Experimental Setup
	3.11 Results and Discussion
		3.11.1 Computer simulations
		3.11.2 Experiment
	3.12 Conclusion
Chapter 4 Capturing Finger Movements
	4.1 Introduction
	4.2 System Overview
	4.3 Accuracy Improvement of Total Active Movement and Proximal Interphalangeal Joint Angles
	4.4 Simulation
	4.5 Trial Procedure
	4.6 Results
		4.6.1 Concurrence validity
		4.6.2 Internal reliability
		4.6.3 Time efficiency
	4.7 Discussions
	4.8 Approaching Finger Movement with a New Perspective
	4.9 Reachable Space
	4.10 Boundary of the Reachable Space
	4.11 Area of the Reachable Space
	4.12 Experiments
	4.13 Results and Discussion
	4.14 Conclusion and Future Work
Chapter 5 Non‐contact Measurement of Respiratory Function via Doppler Radar
	5.1 Introduction
	5.2 Fundamental Operation of Microwave Doppler Radar
		5.2.1 Velocity and frequency
		5.2.2 Correction of I/Q amplitude and phase imbalance
	5.3 Signal Processing Approach
		5.3.1 Respiration rate
		5.3.2 Extracting respiratory signatures
		5.3.3 Low‐pass filtering (LPF)
		5.3.4 Discrete wavelet transform
	5.4 Common Data Acquisitions Setup
	5.5 Capturing the Dynamics of Respiration
		5.5.1 Normal breathing
		5.5.2 Fast breathing
		5.5.3 Slow inhalation–fast exhalation
		5.5.4 Fast inhalation–slow exhalation
		5.5.5 Capturing abnormal breathing patterns
		5.5.6 Breathing component decomposition, analysis and classification
	5.6 Capturing Special Breathing Patterns
		5.6.1 Correlation of radar signal with spirometer in tidal volume estimations
		5.6.2 Experiment setup
		5.6.3 Results
		5.6.4 Motion signature from Doppler radar
		5.6.5 Measurement of volume in (inhalation) and volume out (exhalation)
	5.7 Removal of Motion Artefacts from Doppler Radar‐based Respiratory Measurements
		5.7.1 Experimental verification
		5.7.2 Results and discussion
		5.7.3 Summary
	5.8 Separation of Doppler Radar‐based Respiratory Signatures
		5.8.1 Respiration sensing using Doppler radar
		5.8.2 Signal processing source separation (ICA)
		5.8.3 Experiment protocol for real data sensing
		5.8.4 Two simulated respiratory sources
		5.8.5 Experiment involving real subjects
		5.8.6 Separation of hand motion
		5.8.7 Conclusion
Chapter 6 Appendix
	6.1 Static Estimators
		6.1.1 Least‐squares estimation
		6.1.2 Maximum likelihood estimation
	6.2 Model‐based Estimators
		6.2.1 Kalman filter (KF)
	6.3 Particle Filter
		6.3.1 Robust filtering with linear measurements
		6.3.2 Constrained optimisation
Bibliography
Index
EULA