Muutke küpsiste eelistusi

PPG Signal Analysis: An Introduction Using MATLAB® [Kõva köide]

5.00/5 (4 hinnangut Goodreads-ist)
  • Formaat: Hardback, 270 pages, kõrgus x laius: 234x156 mm, kaal: 721 g, 1 Tables, color; 50 Illustrations, color
  • Ilmumisaeg: 15-Dec-2020
  • Kirjastus: CRC Press
  • ISBN-10: 1138049719
  • ISBN-13: 9781138049710
Teised raamatud teemal:
PPG Signal Analysis: An Introduction Using MATLAB®Suurem pilt
  • Formaat: Hardback, 270 pages, kõrgus x laius: 234x156 mm, kaal: 721 g, 1 Tables, color; 50 Illustrations, color
  • Ilmumisaeg: 15-Dec-2020
  • Kirjastus: CRC Press
  • ISBN-10: 1138049719
  • ISBN-13: 9781138049710
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781138049710.html
Märksõnad:

This book serves as a current resource for Photoplethysmogram (PPG) signal analysis using MATLAB®. This technology is critical in the evaluation of medical and diagnostic data utilized in mobile devices. Information and methodologies outlined in the text can be used to learn the empirical and experimental process (including data collection, data analysis, feature extractions, and more) from inception to conclusion. This book also discusses how introduced methodologies can be used and applied as tools that will teach the user how to validate, test, and simulate developed algorithms before implementing and deploying the algorithms on wearable, battery-driven, or point-of-care devices.

Arvustused

"This is excellent!! Really. PPG Analysis - Introduction Using Matlab provides a unique and relatively comprehensive introduction to photoplethysmogram in the context of digital health. This is a fun, hands-on introduction to the important emerging field of PPG. This book will help inspire a new generation to join us in finding out how to use PPG for multiple digital health applications." - Maxime Cannesson, Professor of Clinical Anesthesiology, UCLA, USA

"A wonderful comprehensive hands-on reference for students and researchers doing practical PPG signal analysis. Although the code examples are in MATLAB, the book content is useful for researchers working with any math package." Richard Ribon Fletcher, Mobile Technology Group, Massachusetts Institute of Technology, USA

"It is a very informative book that is truly needed in the field. It was clearly taught from an engineering perspective. By reading this book the student is given a very real opportunity to make significant discoveries. This book would work well as a supplementary source for an introductory course in computational analysis of PPG signals for undergraduate and postgraduate students." - Kirk Shelley, Professor Emeritus of Anesthesiology, Yale University, USA

"I enjoyed reading this book it encompasses the fascinating field of photoplethysmography and modern analysis techniques using clear Matlab focussed examples. It is a great contribution to the field. I think this book should be given a privileged spot on the bookshelf of every student and researcher in the cardiovascular, vascular optics and digital health sciences." -- Dr John Allen, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK

List of Figures and Tables
xvii
Preface xxi
Acknowledgments xxiii
The Author xxv
How to Use This Book xxvii
Chapter 1 Math Foundations
1(26)
1.1 Learning Objectives
1(1)
1.2 Scalars
1(4)
1.2.1 Scalar Mathematical Operations
1(4)
1.2.2 Assigning Scalar Values
5(1)
1.3 Vectors
5(5)
1.3.1 Vector Mathematical Operations
5(3)
1.3.2 Assigning Vector Elements
8(1)
1.3.3 Assigning Vector Elements Using a Function
8(1)
1.3.4 Assigning Vector Elements Using a Colon (:)
9(1)
1.3.5 Addressing Vector Elements
9(1)
1.3.6 Increasing the Vector Size
10(1)
1.4 Matrices
10(4)
1.4.1 Matrix Mathematical Operations
10(3)
1.4.2 Assigning Matrix Elements
13(1)
1.4.3 Assigning Matrix Elements Using a Function
13(1)
1.4.4 Addressing Matrix Elements
14(1)
1.5 Relational Operators
14(1)
1.6 Nan
15(1)
1.7 Strings
16(3)
1.8 Structures
19(2)
1.9 Cell
21(2)
1.10 Import/Export Data
23(2)
1.11 Workspace User Input
25(2)
Chapter 2 Photoplethysmogram Signals
27(26)
2.1 Learning Objectives
27(1)
2.2 Background
27(2)
2.3 Oxygen Transport
29(1)
2.4 Terminologies And Acronyms
29(6)
2.4.1 DVP
29(1)
2.4.2 PTG
30(1)
2.4.3 SDPTG
31(1)
2.4.4 APG
32(1)
2.4.5 SDDVP
33(1)
2.4.6 Terminology Selection and Search Strategy
33(1)
2.4.7 Standard Acronyms
34(1)
2.5 Why Ppg Signal?
35(1)
2.6 Plethysmography Types
36(2)
2.7 Measuring Sites
38(1)
2.8 Modes Of Ppg Measurement
39(3)
2.8.1 Transmissive Mode
40(1)
2.8.2 Reflective Mode
41(1)
2.9 Calculation Of Oxygen Saturation
42(1)
2.10 Simulation Of Ppg Signal Using Sinusoids
42(2)
2.11 Simulation Of Ppg Signal Using Two Gaussian Functions
44(4)
2.12 Ppg Sensors
48(2)
2.12.1 Probe-Based PPG Signals
48(1)
2.12.2 Video-Based PPG Signals
49(1)
2.13 Current Challenges
50(2)
2.13.1 Powerline Interference
50(1)
2.13.2 Sudden Amplitude Change
50(1)
2.13.3 Motion Artifact
50(1)
2.13.4 Multi-Parameter Systems
51(1)
2.13.5 Research Design
52(1)
2.14 Summary
52(1)
Chapter 3 Visualization of PPG Signals
53(20)
3.1 Learning Objectives
53(1)
3.2 Plot
54(1)
3.3 Bar
54(1)
3.4 Area
55(2)
3.4.1 Histogram
56(1)
3.5 Periodogram
57(1)
3.6 Spectrogram
57(2)
3.6.1 Wavelets
58(1)
3.7 Eventogram
59(3)
3.8 Discussion
62(9)
3.9 Summary
71(2)
Chapter 4 Pre-processing of PPG Signals
73(24)
4.1 Learning Objectives
73(1)
4.2 Filter Types
73(10)
4.2.1 Moving Average (MA) Filter
75(3)
4.2.2 Butterworth Filter (Butter)
78(1)
4.2.3 Chebyshev Filter (Cheby I and Cheby II)
79(2)
4.2.4 Elliptic Filter (Ellip)
81(1)
4.2.5 General Comment
82(1)
4.3 Filter Design
83(6)
4.3.1 Low-Pass Filter
83(2)
4.3.2 High-Pass Filter
85(1)
4.3.3 Band-Pass Filter
86(1)
4.3.4 Band-Stop Filter
87(2)
4.4 Convolution
89(3)
4.4.1 Improving PPG Beat Quality
89(1)
4.4.2 Filtering PPG Signal
90(2)
4.5 Cross-Correlation
92(4)
4.5.1 Filtering One PPG Beat
92(1)
4.5.2 Filtering PPG Signal Quality
93(3)
4.6 Summary
96(1)
Chapter 5 Signal Quality Assessment
97(10)
5.1 Learning Objectives
97(1)
5.2 Introduction
97(1)
5.3 Annotation
98(2)
5.4 Signal Quality Indices
100(4)
5.4.1 Perfusion (PSQI)
101(1)
5.4.2 Skewness (SSQI)
101(1)
5.4.3 Kurtosis (KSQI)
101(1)
5.4.4 Entropy (ESQI)
102(1)
5.4.5 Zero Crossing Rate (ZSQI)
102(1)
5.4.6 Signal-to-Noise Ratio (NSQI)
103(1)
5.4.7 Matching Systolic Detectors (MSQI)
103(1)
5.4.8 Relative Power (RSQI)
104(1)
5.5 Summary
104(3)
Chapter 6 PPG Feature Extraction
107(34)
6.1 Learning Objectives
107(1)
6.2 Overview Of Ppg Features
107(1)
6.3 Features Of Ppg Waveforms
107(5)
6.3.1 Systolic Amplitude
108(1)
6.3.2 Pulse Width
108(1)
6.3.3 Pulse Area
109(1)
6.3.4 Peak-to-Peak Interval
110(1)
6.3.5 Pulse Interval
110(1)
6.3.6 Augmentation Index
110(1)
6.3.7 Large Artery Stiffness Index
111(1)
6.4 Features Of Vpg Signals
112(1)
6.4.1 Diastolic Point
112(1)
6.4.2 ΔT Calculation
113(1)
6.4.3 Crest Time Calculation
113(1)
6.5 Features Of Apg Signals
113(27)
6.5.1 a, b, c, d, and e Waves
113(1)
6.5.2 Ratio b/a Index
114(1)
6.5.3 Ratio c/a Index
115(1)
6.5.4 Ratio d/a Index
115(1)
6.5.5 Ratio e/a Index
115(1)
6.5.6 Ratio (b - c - d - e)/a Index
115(1)
6.5.7 Ratio (b - e)/a Index
116(1)
6.5.8 Ratio (b - c - d)/a Index
116(1)
6.5.9 Ratio (c + d - b)/a Index
116(1)
6.5.10 aa Interval
116(1)
6.5.11 APG Beat Waveform
117(1)
6.5.12 Segment of APG Signal
117(1)
6.5.13 Chaos Attractor
117(1)
6.5.14 MATLAB Functions for Features Extraction
117(3)
6.5.15 MATLAB Code for Extracting 125 PPG Features
120(1)
6.5.15.1 Time Span
120(1)
6.5.15.2 Features of PPG Amplitude
120(1)
6.5.15.3 Features of VPG and APG
121(1)
6.5.15.4 Waveform Area
121(1)
6.5.15.5 Power Area
121(1)
6.5.15.6 Ratio
122(1)
6.5.15.7 Slope
122(1)
6.5.15.8 Code for PPG Feature Calculation
122(8)
6.5.15.9 Heart Rate Variability
130(1)
6.5.15.10 Time Domain Methods
131(1)
6.5.15.11 Frequency Domain Methods
132(2)
6.5.16 Nonlinear Methods
134(1)
6.5.16.1 Poincare Plot
135(1)
6.5.16.2 Approximate Entropy and Sample Entropy
136(3)
6.5.17 Discussion
139(1)
6.6 Summary
140(1)
Chapter 7 A Generic Method for Event Detection
141(24)
7.1 Learning Objectives
141(1)
7.2 Introduction
141(2)
7.3 Data Used
143(1)
7.4 Terma Framework
144(6)
7.4.1 Prior Knowledge
145(1)
7.4.2 Bandpass Filter
146(1)
7.4.3 Signal Enhancement
147(1)
7.4.4 Generating Blocks of Interest
147(1)
7.4.5 Thresholding
148(2)
7.4.6 Detecting Event Peak
150(1)
7.5 Results
150(2)
7.5.1 Training Results
150(1)
7.5.2 Testing
151(1)
7.6 Discussion
152(10)
7.6.1 Frequency Band Choice
153(1)
7.6.2 Window Size Choice
153(3)
7.6.3 Offset β Choice
156(1)
7.6.4 Battery-Driven Devices
157(2)
7.6.5 Optimization Step
159(1)
7.6.5.1 Exhaustive Search
160(1)
7.6.5.2 Gradient-Based Search
161(1)
7.6.5.3 Parallel Execution
162(1)
7.7 Significance Of Terma
162(2)
7.8 Summary
164(1)
Chapter 8 Feature Selection
165(30)
8.1 Learning Objectives
165(1)
8.2 Feature Normalization
165(2)
8.2.1 Linear Normalization
166(1)
8.2.2 Nonlinear Normalization
167(1)
8.3 Criteria For Selection And Evaluation
167(18)
8.3.1 Independent Student's t-Test
167(3)
8.3.2 Dependent Samples (Paired) t-Test
170(1)
8.3.3 Receiver Operating Characteristic Curve
170(2)
8.3.4 Analysis of Variance (ANOVA)
172(6)
8.3.5 Fisher's Measure
178(1)
8.3.6 Divergence Measure
179(1)
8.3.7 Bhattacharyya's Measure
180(1)
8.3.8 Scatter Measure
181(4)
8.4 Optimal Feature(S)
185(4)
8.4.1 Individual Feature Selection
186(3)
8.5 Search Method
189(3)
8.5.1 Optimal Search
189(2)
8.5.2 Suboptimal Search
191(1)
8.6 Summary
192(3)
Chapter 9 Identifying Adverse Events
195(28)
9.1 Learning Objectives
195(1)
9.2 Minimum Distance Classifier
195(3)
9.3 Bayes Classifier
198(2)
9.4 Competitive Neural Network
200(3)
9.5 Discriminant Analysis
203(4)
9.6 Other Classifiers
207(3)
9.7 Classification Example Using Classical Machine Learning Methods
210(3)
9.8 Classification Example Using Deep Learning
213(3)
9.9 Effectiveness Evaluation
216(5)
9.9.1 K-Fold Cross Validation
216(1)
9.9.2 Class Imbalance
217(1)
9.9.3 Confusion Matrix
217(2)
9.9.4 Sensitivity versus Specificity
219(2)
9.10 Summary
221(2)
Chapter 10 Application of PPG to Global Health
223(16)
10.1 Learning Objectives
223(1)
10.2 Introduction
224(1)
10.3 Overview
225(1)
10.4 Simplicity
226(1)
10.5 Mining
227(1)
10.6 Connecting
227(1)
10.7 Reliability
228(1)
10.8 Affordability
229(1)
10.9 Scalability
229(1)
10.10 Noncommunicable Disease Case Studies
230(6)
10.10.1 Case I: Detection of Heat Stress in a Changing Climate
230(1)
10.10.1.1 Simplicity
231(1)
10.10.1.2 Mining
232(1)
10.10.1.3 Connection
232(1)
10.10.1.4 Reliability
232(1)
10.10.1.5 Affordability
233(1)
10.10.1.6 Scalability
233(1)
10.10.2 Case II: Prediction of Adverse Outcomes related to Preeclampsia using SpO2
233(1)
10.10.2.1 Simplicity
233(1)
10.10.2.2 Mining
234(1)
10.10.2.3 Connection
234(1)
10.10.2.4 Affordability
234(1)
10.10.2.5 Scalability
235(1)
10.10.3 Case III: Hypertension Risk Stratification
235(1)
10.10.3.1 Simplicity
235(1)
10.10.3.2 Mining
235(1)
10.10.3.3 Connection
235(1)
10.10.3.4 Affordability
236(1)
10.10.3.5 Scalability
236(1)
10.11 User Performance
236(2)
10.12 Summary
238(1)
Chapter 11 Available PPG Databases
239(8)
11.1 Fingertip Ppg From Hypertensive Subjects
239(1)
11.2 Fingertip Ppg From An Intensive Care Unit
240(1)
11.3 Wrist Ppg During Exercise
241(2)
11.4 Fingertip Ppg And Respiration
243(2)
11.4.1 The University of Queensland Vital Signs Dataset
243(1)
11.4.2 BioSec.Lab PPG Dataset
244(1)
11.4.3 Vortal Dataset
244(1)
11.5 Summary
245(2)
References 247(14)
Index 261
Dr. Mohamed (Moe) Elgendi is currently a Senior Postdoctoral Fellow at UBC's Department of Obstetrics and Gynecology, and Adjunct Professor at UBC's Department of Electrical and Computer Engineering, a Senior Member at IEEE, and a Senior fellow at Howard Brain Sciences Foundation. In addition to his 10+ years of experience in the field of data analysis, he received training on Leadership in Education from MIT. Dr. Elgendi's expertise in the areas of digital health, data science, artificial intelligence, and visualization includes his work in Global Health with the PRE-EMPT Initiative (funded by the Bill and Melinda Gates Foundation), the Institute for Media Innovation at Nanyang Technological University (Singapore), and Alberta's Stollery Children's Hospital (Canada). Dr. Elgendi specializes in bridging the areas of engineering, computer science, psychology, and medicine for knowledge translation.