Mohamed Elsayed AbdelRaouf Elsayed

Body and Visual Sensor Fusion for GAIT Impairment Analysis in Ubiquitous Healthcare Systems Mohamed Elsayed AbdelRaouf Elsayed - 2011 - 122 p. ill. 21 cm.

Supervisor: Mohamed A. El-Helw

Thesis (M.A.)—Nile University, Egypt, 2011 .

"Includes bibliographical references"

Contents:
CHAPTER ONE ............................................................................................................... 19
1. INTRODUCTION..................................................................................................... 19
1.1. Introduction .................................................................................................. 19
1.2. Challenges ..................................................................................................... 22
1.3. Thesis Contribution ....................................................................................... 23
CHAPTER TWO .............................................................................................................. 26
2. PERVASIVE HEALTHCARE MONITORING ................................................................ 26
2.1. Introduction .................................................................................................. 26
2.2. Ubiquitous Monitoring Technologies ............................................................. 27
2.2.1. Body Sensor Networks ........................................................................... 27
2.2.2. Visual Sensor Networks ......................................................................... 30
2.3. Sensor Design and Hardware Architectures ................................................... 33
2.3.1. BSN Design Goals ................................................................................... 33
2.3.2. BSNs Hardware Architecture .................................................................. 35
2.3.3. VSNs Design Goals ................................................................................. 38
2.3.4. VSNs Hardware Architecture.................................................................. 38
2.4. Motion Analysis Applications ......................................................................... 41
2.5. Conclusions ................................................................................................... 44
CHAPTER THREE ........................................................................................................... 45
3. BODY AND VISUAL MOTION FEATURES EXTRACTION AND FUSION ........................ 45
3.1. Introduction ..................................................................................................
3.2. Body Motion Feature Computation ............................................................... 45
3.3. Visual Feature Extraction ............................................................................... 49
3.3.1. Visual Data Enhancement ............................................................................ 49
Background Modeling and Foreground Detection................................... 49
Shadow Detection and Removal ............................................................. 53
3.3.2. Blob Skeletonization .................................................................................... 56
Moore Neighborhood algorithm ............................................................. 57
3.3.3. Visual Metrics .............................................................................................. 60
3.4. Data Fusion ................................................................................................... 64
3.5. Conclusions ................................................................................................... 67
CHAPTER FOUR ............................................................................................................. 68
4. GAIT IMPAIRMENT CLASSIFICATION AND ANALYSIS .............................................. 68
4.1. Introduction .................................................................................................. 68
4.2. Classification Techniques ............................................................................... 68
4.2.1. Multilayer Perceptron ................................................................................. 68
4.2.2. Naïve Bayes Classifier .................................................................................. 71
4.2.3. K-Nearest Neighbor ..................................................................................... 73
4.2.4. Decision Tree............................................................................................... 75
4.3. Change Analysis Technique............................................................................ 76
4.4. Conclusion ..................................................................................................... 79
CHAPTER FIVE ............................................................................................................... 80
5. EXPERIMENTS AND RESULTS ................................................................................. 80
5.1. Introduction .................................................................................................. 80
5.2. Sensors Placement and Installation ............................................................... 81
5.2.1. BSN Sensor Placement and Orientation ....................................................... 81
5.2.2. VSN Nodes Deployment .............................................................................. 82
5.3. Dataset Collection ......................................................................................... 85
5.4. Cross-Validation Experiment.......................................................................... 87
5.5. Gait Classification Experiment ....................................................................... 91
5.4.1. Multiple Layer Preceptron ...........................................................................
5.4.2. Naïve Bayes Classifier .................................................................................. 92
5.4.3. K-Nearest Neighbors ................................................................................... 93
5.4.4. Decision Tree............................................................................................... 95
5.4.5. Error Percentage ......................................................................................... 96
5.6. Gait Impairment Quantification using Change Analysis Technique ................. 97
5.7. Conclusion ................................................................................................... 101
CHAPTER SIX ............................................................................................................... 102
6. CONCLUSIONS AND FUTURE WORK ....................................................................

Abstract:
Human motion analysis provides a valuable solution for monitoring the
wellbeing of the elderly, quantifying post-operative patient recovery and monitoring
the progression of neurodegenerative diseases such as Parkinson’s. The
development of accurate motion analysis models, however, requires the integration
of multi-sensing modalities and the utilization of appropriate data analysis
techniques. This work describes a robust framework that is monitor and quantifies
the patients’ gait impairment by integrating information captured by two different
sensing modalities which are body and visual sensor networks. Real-time target
extraction is applied and a skeletonization procedure is subsequently carried out to
quantify the internal motion of moving target and computes two metrics,
spatiotemporal cyclic motion between leg segments and head trajectory, for each
vision node. Extracted motion metrics from multiple vision nodes and accelerometer
information from a wearable body sensor are then fused at the feature level. Then
the combined features used to classify target’s walking gait as being normal or
abnormal (limping) by using Cross-Validation and several classification techniques
such as Multiple Layer Preceptron, Naïve Bayes Classifier, K-Nearest Neighbors,
and Decision Tree. Furthermore the proposed framework is capable of measuring
gait impairment and recovery of postoperative patients in smart healthcare
environments by using Change Analysis algorithm. The potential value of the
proposed framework for patient monitoring is demonstrated and the results obtained
from practical experiments are described.

Text in English, abstracts in English.

Subjects--Topical Terms:
Informatics-IFM

Index Terms--Genre/Form:
Dissertation, Academic

Dewey Class. No.: 610