Computational Diffusion MRI: International MICCAI Workshop, Lima, Peru, October 2020 (Mathematics and Visualization)

Programme Committee
Preface
Contents
Diffusion MRI Signal Acquisition
Image Reconstruction from Accelerated Slice-Interleaved Diffusion Encoding Data
	1 Introduction
	2 Methods
		2.1 SIDE Acquisition
		2.2 Reconstruction
		2.3 Optimization
	3 Experiments
		3.1 Materials
		3.2 Results
	4 Conclusion
	References
Towards Learned Optimal q-Space Sampling in Diffusion MRI
	1 Introduction
		1.1 Main Contributions
	2 Method
		2.1 Forward Model: Sub-Sampling Layer
		2.2 Reconstruction Model
		2.3 Optimization
	3 Experimental Evaluation
		3.1 Dataset
		3.2 Training Settings
		3.3 Results and Discussion
	4 Conclusion
	5 Supplementary Materials
	References
A Signal Peak Separation Indexpg  for Axisymmetric B-Tensor Encoding
	1 Introduction
	2 Theory
		2.1 A Toy Model of Fascicle Crossing Under B-Tensor Encoding
		2.2 The Signal Peak Separation Index
	3 Methods
	4 Results
	5 Discussion and Conclusion
	References
Orientation Processing: Tractography and Visualization
Improving Tractography Accuracy Using Dynamic Filtering
	1 Introduction
	2 Materials and Methods
		2.1 Initial Set of Streamlines
		2.2 Parametric Representation of the Streamlines
		2.3 Optimization
		2.4 Data and Experiments
	3 Results and Discussion
	4 Conclusions
	References
Diffeomorphic Alignment of Along-Tract Diffusion Profiles from Tractography
	1 Introduction
	2 Alignment of Along-Tract Diffusion Measure Profiles
		2.1 Representation
		2.2 Objective Function for Joint Alignment
		2.3 Alternating Minimization for Subject-Level and Tract-Level Alignment
	3 Results
		3.1 Data
		3.2 Along-Tract FA Profiles Before and After Joint Alignment
		3.3 Reduced Coefficient of Variation
		3.4 Subject-Wise Inter-tract Correlations
		3.5 Intraclass Correlation Coefficient for Reliability Across Time Points
	4 Discussion
	References
Direct Reconstruction of Crossing Muscle Fibers in the Human Tongue Using a Deep Neural Network
	1 Introduction
	2 Methods
		2.1 Training Data and Ground Truth
		2.2 Fiber Estimation Network
		2.3 Fiber Estimation Loss
		2.4 Training Procedure
	3 Experiments and Results
		3.1 Quantitative Evaluation on Synthetic Tongue HARDI Data
		3.2 Qualitative Results on Post-mortem Human Tongue Data
	4 Discussion and Conclusions
	References
Learning Anatomical Segmentations  for Tractography from Diffusion MRI
	1 Introduction
	2 Materials and Methods
		2.1 Data
		2.2 Data Representations
		2.3 Architecture
		2.4 Training
		2.5 Tracts
		2.6 Evaluation Criteria
	3 Results and Discussion
		3.1 Evaluation 1: Q-Space Sampling Density
		3.2 Evaluation 2: Input Representations
		3.3 Evaluation 3: Generalization
		3.4 Evaluation 4: Tract Similarity
	4 Conclusion
	References
Diffusion MRI Fiber Orientation Distribution Function Estimation Using Voxel-Wise Spherical U-Net
	1 Introduction
	2 Background and Method
		2.1 Voxel-Wise Spherical U-Net
	3 Dataset
	4 Experiments and Implementation Details
	5 Results and Conclusions
	References
Microstructure Modeling and Representation
Stick Stippling for Joint 3D Visualization of Diffusion MRI Fiber Orientations and Density
	1 Introduction
	2 Methods
		2.1 Diffusion Modeling and the Fixel Representation
		2.2 Fixel Glyph Visualization
	3 Experiments and Results
		3.1 Clinical Data Experiment
		3.2 HCP Experiment
		3.3 RESOLVE Experiment
	4 Discussion and Conclusions
	References
Q-Space Quantitative Diffusion MRI Measures Using a Stretched-Exponential Representation
	1 Introduction
	2 Theory
		2.1 Diffusion MR Signal Representation
		2.2 Q-Space Domain Quantitative Measures
		2.3 Numerical Implementation
		2.4 Optimization of Stretched-Exponential Representation
	3 Materials and Methods
		3.1 Ex Vivo rat brain data
		3.2 In Vivo Human brain data
		3.3 Comparison to the Q-Space Measures from Different Methods
	4 Results and Discussion
	5 Conclusions
	References
Repeatability of Soma and Neurite Metrics in Cortical and Subcortical Grey Matter
	1 Introduction
	2 Methods
		2.1 Image Acquisition and Pre-processing
		2.2 Image Processing and Analysis
	3 Results
	4 Discussion
	References
DW-MRI Microstructure Model  of Models Captured Via Single-Shell Bottleneck Deep Learning
	1 Introduction
	2 Related Work
	3 Data Acquisition
	4 Proposed Method
	5 Results
	6 Discussion
	References
Deep Learning Model Fitting  for Diffusion-Relaxometry: A Comparative Study
	1 Introduction
	2 Methods
		2.1 qMRI Model Fitting with DNNs
		2.2 In Silico Study
		2.3 In Vivo study
	3 Results
	4 Discussion
	5 Conclusion
	References
Pretraining Improves Deep Learning Based Tissue Microstructure Estimation
	1 Introduction
	2 Methods
		2.1 Problem Formulation
		2.2 Signal Generation for Pretraining
		2.3 Backbone Deep Network
		2.4 Pretraining with the Auxiliary Dataset and Fine-Tuning
		2.5 Implementation Details
	3 Results
	4 Discussion
	5 Conclusion
	References
Signal Augmentation and Super Resolution
Enhancing Diffusion Signal Augmentation Using Spherical Convolutions
	1 Introduction
	2 Signal Augmentation
		2.1 Deep Learning Models
		2.2 Spherical Deep Learning Models
		2.3 Material
	3 Evaluation
		3.1 Results
	4 Discussion
	5 Conclusion
	References
Hybrid Graph Convolutional Neural Networks for Super Resolution of DW Images
	1 Introduction
	2 Dataset
	3 Methods
		3.1 Coarse SR Prediction in 3D Grid Structure Space
		3.2 Refinement by GCNN in Diffusion Gradient Space
		3.3 Loss Function
	4 Experiments
	5 Conclusion
	References
Manifold-Aware CycleGAN  for High-Resolution Structural-to-DTI Synthesis
	1 Introduction
	2 Method
		2.1 Log-Euclidean Metric
		2.2 Adversarial Loss
		2.3 Cycle Consistency Loss
		2.4 Manifold-Aware Wasserstein CycleGAN
	3 Experiments
	4 Discussion and Conclusion
	References
Diffusion MRI Applications
Beyond Lesion-Load: Tractometry-Based Metrics for Characterizing White Matter Lesions within Fibre Pathways
	1 Introduction
	2 Theory and Methods
		2.1 Clinical Assessment
		2.2 Acquisition
		2.3 Processing
		2.4 Proposed Metrics
	3 Results
		3.1 Lesion Mapping
		3.2 Volumetric Metrics
		3.3 Tractometry-Based Metrics
	4 Discussion and Conclusion
	References
Multi-modal Brain Age Estimation:  A Comparative Study Confirms the Importance of Microstructure
	1 Introduction
	2 Data and Materials
	3 Methods
		3.1 Brain Age Estimation
		3.2 Associations with IDPs and Non-IDP Variables
	4 Results
		4.1 Brain Age Estimation
		4.2 Association with Brain IDPs
		4.3 Association with Cardiac Variables
	5 Discussion
	References
Longitudinal Parcellation of the Infant Cortex Using Multi-modal Connectome Harmonics
	1 Introduction
	2 Methods
		2.1 Data and Preprocessing
		2.2 Connectivity Matrices
		2.3 Iterative Multi-modal Parcellation Via Connectome Harmonics
		2.4 Optimal Cluster Number Determination
	3 Results
		3.1 Homogeneity
		3.2 Community Detection
	4 Discussion
	5 Conclusion
	References
Automatic Segmentation of Dentate Nuclei for Microstructure Assessment: Example of Application to Temporal Lobe Epilepsy Patients
	1 Introduction
	2 Methods
		2.1 Subjects
		2.2 MRI Protocol
		2.3 DWI Processing
		2.4 DNs Segmentation
		2.5 Post Processing for OPAL and CNN
		2.6 Quantitative Evaluation
		2.7 Comparison of Automatic Methods
		2.8 Clinical Application to TLE Data
	3 Results
		3.1 Comparison of the Three Automatic Methods
		3.2 Application to TLE Dataset
	4 Discussion
	5 Conclusion
	References
Two Parallel Stages Deep Learning Network for Anterior Visual Pathway Segmentation
	1 Introduction
	2 Methods
		2.1 Data Preprocessing
		2.2 Two Parallel Stages Network Architecture
	3 Experiments
		3.1 Dataset
		3.2 Implementation Details
		3.3 Results
	4 Conclusion
	References
Exploring DTI Benchmark Databases Through Visual Analytics
	1 Introduction
	2 Related Work
	3 Use Case
	4 Implementation
	5 Discussion
	6 Conclusions and Future Work
	References
Index