Linked e-resources
Details
Table of Contents
Intro
Preface
Acknowledgement
Author biographies
Kit Windows-Yule
Leonard Nicuşan
Matthew T Herald
Samuel Manger
David Parker
Chapter 0 Using the book
0.1 The 'User'
0.2 The 'Researcher'
0.3 The 'Developer'
0.4 The 'Expert'
Chapter 1 Imaging particulate and multiphase systems
1.1 Particulate and multiphase systems: why do they matter?
1.2 The importance of imaging
1.3 Particle and flow imaging: an overview
References
Chapter 2 The fundamentals of PEPT
2.1 Positron emission…
2.2 …particle tracking
2.2.1 Interactive example: PEPT-an idealised case
PEPT: an idealised case
Monte Carlo line of response generation
Triangulate tracer's location
Spatial error versus number of LoRs used
2.3 A more realistic picture
2.3.1 Issue 1: false coincidences
2.3.2 Issue 2: positron flight
2.3.3 Issue 3: imperfect detectors
2.3.4 Issue 4: finite detection rate
2.3.5 The real picture
2.3.6 Interactive example: sources of error in PEPT
Sources of error in PEPT
Monte Carlo line of response generation
Adding noise: scattered events and spread
Triangulate tracer's location
Spatial error versus noise ratio
2.4 Not just particle tracking
References
Chapter 3 A history of PEPT
3.1 Adding the 'P': from PET to PEPT (origins to 1989)
3.2 MWPC PEPT (1989-1999)
3.3 Out with the old, in with the new (1999-2002)
3.4 Positron cameras of all shapes and sizes (2002-present)
3.5 PEPT elsewhere
3.6 The future-SuperPEPT, MicroPEPT and beyond
References
Chapter 4 Comparison with other techniques
4.1 Quasi-two-dimensional techniques
4.1.1 Particle tracking velocimetry (PTV)
4.1.2 Particle imaging velocimetry (PIV)
4.1.3 Photoelastic imaging
4.1.4 X-ray radiography
4.2 Tomographic techniques
4.2.1 X-ray computed tomography.
4.2.2 Single-photon emission computed tomography (SPECT)
4.2.3 Positron emission tomography (PET)
4.2.4 Electrical capacitance tomography (ECT)
4.2.5 Magnetic resonance imaging (MRI)
4.2.6 Refractive index matched scanning (RIMS)
4.3 Three-dimensional particle-tracking techniques
4.3.1 Radioactive particle tracking (RPT)
4.3.2 Magnetic particle tracking (MPT)
4.4 Non-imaging techniques
Angle of repose (AOR) tests
Tapped density tests-the Carr index and Hausner ratio
Flow through an orifice
Shear testing
Powder rheometry
4.5 Numerical simulation
4.5.1 The discrete element method (DEM)
A simple discrete element method simulation
Define system of differential equations
Numerically solving the ODE: Euler method
Numerically solving the ODE: adaptive integration schemes
4.5.2 Computational fluid dynamics (CFD)
4.5.3 The Monte Carlo method
A simple Monte Carlo simulation
How many random samples?
4.6 Other techniques
References
Chapter 5 Tracers and detectors
5.1 Creating tracers
5.1.1 Introduction-the ideal tracer
5.1.2 Positron-emitting nuclides and direct activation
5.1.3 Indirect activation
5.1.4 Handling and coating
5.2 Detector systems
5.2.1 Introduction
5.2.2 Scintillation detectors
5.2.3 PEPT system geometries
5.2.4 Examples of PEPT systems
5.2.5 Future developments
5.3 Modelling PEPT systems
5.3.1 How it works
5.3.2 Existing GATE models
5.3.3 Modelling a PEPT detector
5.3.4 Defining a PEPT tracer
5.3.5 Recreating an experiment
5.3.6 Studying a PEPT system using GATE
References
Chapter 6 Pre-processing: PEPT data and algorithms
6.1 Understanding PEPT data
6.1.1 Interactive example: PEPT data format
PEPT Data Format
Initialise raw line of response data
Visualising a sample of LoRs.
Temporal resolution?
6.2 Available algorithms
6.2.1 The Birmingham algorithm
6.2.2 Interactive example: the Birmingham algorithm
Interactive PEPT analysis example using the Birmingham method [1]
This Jupyter Notebook
Initialise raw line of response data
Find minimum distance point
Remove the farthest lines of response
Iteratively remove the farthest LoRs and recompute MDP
Complete Birmingham Method code
High-performance Birmingham Method implementation
6.2.3 The line-density method
6.2.4 Interactive example: the line-density method
Interactive PEPT analysis example using the line density algorithm [4]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
Fit 1D Gaussians around the peak
Complete line density method code
6.2.5 The G-means clustering algorithm
6.2.6 Interactive example: the G-means algorithm
Interactive PEPT analysis example using the clustering (G-means) algorithm [6]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
High pass filter
Cluster voxels with G-means
Complete clustering (G-means) algorithm code
Multiple particle tracking
6.2.7 Feature-point identification (FPI)
6.2.8 Interactive example: FPI
Interactive PEPT analysis example using the feature point identification algorithm [17]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
Subtract convolved matrix and blur
Extract voxel peaks
Complete FPI algorithm code
Multiple particle tracking
High-performance FPI algorithm implementation
6.2.9 Spatiotemporal B-spline reconstruction (SBSR)
6.2.10 Voronoi tesselation method
6.2.11 Interactive example: Voronoi tesselation.
Interactive PEPT analysis example using the Voronoi tesselation method [30]
This Jupyter Notebook
Initialise raw line of response data
Discretise the lines of response
Voronoi tesselation
Gather points under consideration (PUCs)
Local filtering based on the local outlier factor
Global filtering
Clustering the remaining PUCs
Extract final tracer locations
Complete Voronoi tesselation algorithm code
6.2.12 The triangulation method
6.2.13 Interactive example: triangulation method
Interactive PEPT analysis example using the triangulation method [37]
This Jupyter Notebook
Initialise raw line of response data
Calculate the LoR distance matrix
Cluster LoRs closer than the tracer radius
Find centroids of clustered LoRs' cutpoints
Complete triangulation method code
6.2.14 PEPT using machine learning (PEPT-ML)
6.2.15 Interactive example: PEPT-ML
Interactive PEPT analysis example using the PEPT-ML algorithm [38]
This Jupyter Notebook
Initialise raw line of response data
Find cutpoints
Cluster cutpoints with HDBSCAN
Compute cluster centres
Complete PEPT-ML algorithm code
Second pass of clustering
Multiple particle tracking
High-performance PEPT-ML algorithm implementation
6.2.16 PEPT using expectation-maximisation (PEPT-EM)
6.2.17 Interactive example: PEPT-EM
Interactive PEPT analysis examples using PEPT-EM [44]
This Jupyter Notebook
Initialise Raw Line of Response Data
Calculate MDP and assign a weight to each LoR
Recalculate MDP with previous weights
Complete PEPT-EM algorithm code
6.2.18 The K-medoids method
6.2.19 Interactive example: K-medoids method
Interactive PEPT analysis example using the K-medoids method [45]
This Jupyter Notebook
Initialise raw line of response data
Find cutpoints.
Filter cutpoints with far nearest neighbors
Cluster filtered cutpoints using K-medoids
Compute clusters' centroids
Complete K-medoids method mode
Multiple particle pracking
6.2.20 The multiple location-allocation algorithm (MLAA)
6.2.21 Interactive example: the multiple location-allocation algorithm (MLAA)
Interactive PEPT analysis example using the multiple location-allocation algorithm (MLAA) [49]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
Voxel global thresholding
The location-allocation algorithm
Complete multiple location-allocation algorithm code
Multiple particle tracking
6.3 From finding tracers to tracking trajectories
6.3.1 Interactive example: the effects of sample size and overlap
Effect of sample size and overlap
Prelude
Initialise raw line of response data
Effect of sample size
Effect of overlap
6.3.2 Trajectory extraction
6.3.3 Interactive example: filtering trajectories
Filtering trajectories
Initialising lines of response
Filtering based on spatial error
Filtering based on nearest neighbours
Filtering using the PEPT Library
6.3.4 Interactive example: separating trajectories
Separating trajectories
Initialising lines of response
PTV-based trajectory separation
Clustering-based trajectory separation
6.4 Horses for courses: comparing algorithm capability for differing tasks
References
Chapter 7 Post-processing: extracting physical information from PEPT data
7.1 Particle trajectories
7.1.1 Single-particle trajectories and their interpretation
7.1.2 Interactive example: plotting single particle trajectories
Tutorial: Using PEPT data to plot single particle trajectories
Setting up
Importing data
Plotting data
Using the pept library.
7.1.3 Interactive example: multiple-particle data.
Preface
Acknowledgement
Author biographies
Kit Windows-Yule
Leonard Nicuşan
Matthew T Herald
Samuel Manger
David Parker
Chapter 0 Using the book
0.1 The 'User'
0.2 The 'Researcher'
0.3 The 'Developer'
0.4 The 'Expert'
Chapter 1 Imaging particulate and multiphase systems
1.1 Particulate and multiphase systems: why do they matter?
1.2 The importance of imaging
1.3 Particle and flow imaging: an overview
References
Chapter 2 The fundamentals of PEPT
2.1 Positron emission…
2.2 …particle tracking
2.2.1 Interactive example: PEPT-an idealised case
PEPT: an idealised case
Monte Carlo line of response generation
Triangulate tracer's location
Spatial error versus number of LoRs used
2.3 A more realistic picture
2.3.1 Issue 1: false coincidences
2.3.2 Issue 2: positron flight
2.3.3 Issue 3: imperfect detectors
2.3.4 Issue 4: finite detection rate
2.3.5 The real picture
2.3.6 Interactive example: sources of error in PEPT
Sources of error in PEPT
Monte Carlo line of response generation
Adding noise: scattered events and spread
Triangulate tracer's location
Spatial error versus noise ratio
2.4 Not just particle tracking
References
Chapter 3 A history of PEPT
3.1 Adding the 'P': from PET to PEPT (origins to 1989)
3.2 MWPC PEPT (1989-1999)
3.3 Out with the old, in with the new (1999-2002)
3.4 Positron cameras of all shapes and sizes (2002-present)
3.5 PEPT elsewhere
3.6 The future-SuperPEPT, MicroPEPT and beyond
References
Chapter 4 Comparison with other techniques
4.1 Quasi-two-dimensional techniques
4.1.1 Particle tracking velocimetry (PTV)
4.1.2 Particle imaging velocimetry (PIV)
4.1.3 Photoelastic imaging
4.1.4 X-ray radiography
4.2 Tomographic techniques
4.2.1 X-ray computed tomography.
4.2.2 Single-photon emission computed tomography (SPECT)
4.2.3 Positron emission tomography (PET)
4.2.4 Electrical capacitance tomography (ECT)
4.2.5 Magnetic resonance imaging (MRI)
4.2.6 Refractive index matched scanning (RIMS)
4.3 Three-dimensional particle-tracking techniques
4.3.1 Radioactive particle tracking (RPT)
4.3.2 Magnetic particle tracking (MPT)
4.4 Non-imaging techniques
Angle of repose (AOR) tests
Tapped density tests-the Carr index and Hausner ratio
Flow through an orifice
Shear testing
Powder rheometry
4.5 Numerical simulation
4.5.1 The discrete element method (DEM)
A simple discrete element method simulation
Define system of differential equations
Numerically solving the ODE: Euler method
Numerically solving the ODE: adaptive integration schemes
4.5.2 Computational fluid dynamics (CFD)
4.5.3 The Monte Carlo method
A simple Monte Carlo simulation
How many random samples?
4.6 Other techniques
References
Chapter 5 Tracers and detectors
5.1 Creating tracers
5.1.1 Introduction-the ideal tracer
5.1.2 Positron-emitting nuclides and direct activation
5.1.3 Indirect activation
5.1.4 Handling and coating
5.2 Detector systems
5.2.1 Introduction
5.2.2 Scintillation detectors
5.2.3 PEPT system geometries
5.2.4 Examples of PEPT systems
5.2.5 Future developments
5.3 Modelling PEPT systems
5.3.1 How it works
5.3.2 Existing GATE models
5.3.3 Modelling a PEPT detector
5.3.4 Defining a PEPT tracer
5.3.5 Recreating an experiment
5.3.6 Studying a PEPT system using GATE
References
Chapter 6 Pre-processing: PEPT data and algorithms
6.1 Understanding PEPT data
6.1.1 Interactive example: PEPT data format
PEPT Data Format
Initialise raw line of response data
Visualising a sample of LoRs.
Temporal resolution?
6.2 Available algorithms
6.2.1 The Birmingham algorithm
6.2.2 Interactive example: the Birmingham algorithm
Interactive PEPT analysis example using the Birmingham method [1]
This Jupyter Notebook
Initialise raw line of response data
Find minimum distance point
Remove the farthest lines of response
Iteratively remove the farthest LoRs and recompute MDP
Complete Birmingham Method code
High-performance Birmingham Method implementation
6.2.3 The line-density method
6.2.4 Interactive example: the line-density method
Interactive PEPT analysis example using the line density algorithm [4]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
Fit 1D Gaussians around the peak
Complete line density method code
6.2.5 The G-means clustering algorithm
6.2.6 Interactive example: the G-means algorithm
Interactive PEPT analysis example using the clustering (G-means) algorithm [6]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
High pass filter
Cluster voxels with G-means
Complete clustering (G-means) algorithm code
Multiple particle tracking
6.2.7 Feature-point identification (FPI)
6.2.8 Interactive example: FPI
Interactive PEPT analysis example using the feature point identification algorithm [17]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
Subtract convolved matrix and blur
Extract voxel peaks
Complete FPI algorithm code
Multiple particle tracking
High-performance FPI algorithm implementation
6.2.9 Spatiotemporal B-spline reconstruction (SBSR)
6.2.10 Voronoi tesselation method
6.2.11 Interactive example: Voronoi tesselation.
Interactive PEPT analysis example using the Voronoi tesselation method [30]
This Jupyter Notebook
Initialise raw line of response data
Discretise the lines of response
Voronoi tesselation
Gather points under consideration (PUCs)
Local filtering based on the local outlier factor
Global filtering
Clustering the remaining PUCs
Extract final tracer locations
Complete Voronoi tesselation algorithm code
6.2.12 The triangulation method
6.2.13 Interactive example: triangulation method
Interactive PEPT analysis example using the triangulation method [37]
This Jupyter Notebook
Initialise raw line of response data
Calculate the LoR distance matrix
Cluster LoRs closer than the tracer radius
Find centroids of clustered LoRs' cutpoints
Complete triangulation method code
6.2.14 PEPT using machine learning (PEPT-ML)
6.2.15 Interactive example: PEPT-ML
Interactive PEPT analysis example using the PEPT-ML algorithm [38]
This Jupyter Notebook
Initialise raw line of response data
Find cutpoints
Cluster cutpoints with HDBSCAN
Compute cluster centres
Complete PEPT-ML algorithm code
Second pass of clustering
Multiple particle tracking
High-performance PEPT-ML algorithm implementation
6.2.16 PEPT using expectation-maximisation (PEPT-EM)
6.2.17 Interactive example: PEPT-EM
Interactive PEPT analysis examples using PEPT-EM [44]
This Jupyter Notebook
Initialise Raw Line of Response Data
Calculate MDP and assign a weight to each LoR
Recalculate MDP with previous weights
Complete PEPT-EM algorithm code
6.2.18 The K-medoids method
6.2.19 Interactive example: K-medoids method
Interactive PEPT analysis example using the K-medoids method [45]
This Jupyter Notebook
Initialise raw line of response data
Find cutpoints.
Filter cutpoints with far nearest neighbors
Cluster filtered cutpoints using K-medoids
Compute clusters' centroids
Complete K-medoids method mode
Multiple particle pracking
6.2.20 The multiple location-allocation algorithm (MLAA)
6.2.21 Interactive example: the multiple location-allocation algorithm (MLAA)
Interactive PEPT analysis example using the multiple location-allocation algorithm (MLAA) [49]
This Jupyter Notebook
Initialise raw line of response data
Voxelise the lines of response
Voxel global thresholding
The location-allocation algorithm
Complete multiple location-allocation algorithm code
Multiple particle tracking
6.3 From finding tracers to tracking trajectories
6.3.1 Interactive example: the effects of sample size and overlap
Effect of sample size and overlap
Prelude
Initialise raw line of response data
Effect of sample size
Effect of overlap
6.3.2 Trajectory extraction
6.3.3 Interactive example: filtering trajectories
Filtering trajectories
Initialising lines of response
Filtering based on spatial error
Filtering based on nearest neighbours
Filtering using the PEPT Library
6.3.4 Interactive example: separating trajectories
Separating trajectories
Initialising lines of response
PTV-based trajectory separation
Clustering-based trajectory separation
6.4 Horses for courses: comparing algorithm capability for differing tasks
References
Chapter 7 Post-processing: extracting physical information from PEPT data
7.1 Particle trajectories
7.1.1 Single-particle trajectories and their interpretation
7.1.2 Interactive example: plotting single particle trajectories
Tutorial: Using PEPT data to plot single particle trajectories
Setting up
Importing data
Plotting data
Using the pept library.
7.1.3 Interactive example: multiple-particle data.