Computational Diffusion MRI: 13th International Workshop, CDMRI 2022 Held in Conjunction with MICCAI 2022 Singapore, Singapore, September 22, 2022 Proceedings

Description

This book constitutes the proceedings of the International Workshop on Computational Diffusion MRI, CDMRI 2022, which was held 22 September 2022, in conjunction with MICCAI 2022.  The 12 full papers included were carefully reviewed and selected for inclusion in the book.  The papers were organized in topical sections as follows: Data processing, Signal representations, Tractography and WM pathways. Preface Organization Contents Data Preprocessing Slice Estimation in Diffusion MRI of Neonatal and Fetal Brains in Image and Spherical Harmonics Domains Using Autoencoders 1 Introduction 2 Methodology 2.1 Materials 2.2 Model 3 Results 3.1 DWI Assessment 3.2 FA and MD in Newborns 3.3 Qualitative Results of FA and MD in Fetuses 4 Conclusion References Super-Resolution of Manifold-Valued Diffusion MRI Refined by Multi-modal Imaging 1 Introduction 2 Background and Methods 2.1 Background 2.2 Proposed Method 3 Experiments 3.1 Data 3.2 Implementation and Training Details 3.3 Models and Evaluation Metrics 4 Results 4.1 Proposed Model Performance 4.2 Model Ablation Results 5 Discussion References .28em plus .1em minus .1emLossy Compression of Multidimensional Medical Images Using Sinusoidal Activation Networks: An Evaluation Study 1 Introduction 2 Methodology 3 Results 4 Discussion and Conclusion References Correction of Susceptibility Distortion in EPI: A Semi-supervised Approach with Deep Learning 1 Introduction 2 Background 2.1 Distortion Model 2.2 Distortion Correction Using Image Registration 3 Method 3.1 Model Architecture 3.2 Models 4 Evaluations 4.1 Datasets 4.2 Models 4.3 Assessment Metrics 5 Results 6 Discussion 7 Conclusion References The Impact of Susceptibility Distortion Correction Protocols on Adolescent Diffusion MRI Measures 1 Introduction 2 Methods 2.1 Study Participants and MRI Data 2.2 dMRI Preprocessing and Subsampling 2.3 dMRI Models and Regional Measures 2.4 Statistics 3 Results 3.1 DTI and NODDI Map Comparisons 3.2 DTI and NODDI Fit Evaluations 3.3 GAM Age Associations 4 Discussion References Signal Representations Diffusion MRI Fibre Orientation Distribution Inpainting 1 Introduction 2 Methods 2.1 The Human Connectome Project Dataset 2.2 Data Preprocessing 2.3 3D FOD Inpainting Framework 2.4 Feature Encoding Stage 2.5 Order-Wise Coefficient Decoders 2.6 Implementation Details 3 Experimental Results 3.1 Inpainting Quality Analysis 3.2 Connectome Matrix Analysis 4 Conclusion References Fitting a Directional Microstructure Model to Diffusion-Relaxation MRI Data with Self-supervised Machine Learning 1 Introduction 2 Methods 2.1 Microstructure Model 2.2 Combined T1-Diffusion in Vivo Data 2.3 Simulated Data 2.4 Non-linear Least Squares Fitting 2.5 Self-supervised Model Fitting 3 Results 3.1 Simulated Data 3.2 Real Data 4 Discussion 5 Conclusion References Stepwise Stochastic Dictionary Adaptation Improves Microstructure Reconstruction with Orientation Distribution Function Fingerprinting 1 Introduction 2 Methods 2.1 Biophysical Diffusion Model 2.2 Orientation Distribution Function Fingerprinting 2.3 Stepwise Stochastic Adaptation of a Dictionary 2.4 Data 2.5 Evaluation 3 Results 4 Discussion 5 Conclusions References How Can Spherical CNNs Benefit ML-Based Diffusion MRI Parameter Estimation? 1 Introduction 2 ML Solutions to the dMRI Parameter Estimation Problem and the Theoretical Benefits of S-CNNs 2.1 Fully-Connected Networks 2.2 Spherical CNNs 3 Experiments 3.1 Experiment 1 3.2 Experiment 2 4 Results and Discussion 4.1 For Experiment 1 4.2 For Experiment 2 5 Conclusion References Tractography and WM Pathways DC2U-Net: Tract Segmentation in Brain White Matter Using Dense Criss-Cross U-Net 1 Introduction 2 Methods 2.1 Dense Criss-Cross U-Net 2.2 Dense Criss-Cross Attention (DCCA) Block 2.3 Deeply Supervised Loss Function 3 Experiments 3.1 Dataset and Implementation Details 3.2 Results 3.3 Ablation Analysis of DC2U-Net 4 Conclusion References Clustering in Tractography Using Autoencoders (CINTA) 1 Introduction 1.1 Related Work 2 Material and Methods 3 Experiments 4 Results 5 Discussion 6 Conclusion A Appendix A.1 Misclassified Streamlines A.2 Time Computational Requirements References Tractometric Coherence of Fiber Bundles in DTI 1 Introduction 2 Theory 3 Experiments 4 Discussion References Author Index