Machine Learning in Medical Imaging

Name: Machine Learning in Medical Imaging | 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16, 2018, Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16, 2018, Proceedings

Yinghuan Shi Heung-Il Suk Mingxia Liu(Editor)

Springer (Publisher)

Published on 14. September 2018

XII, 409 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-00919-9 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
Contents
Developing Novel Weighted Correlation Kernels for Convolutional Neural Networks to Extract Hierarchical Functional Connectivities from fMRI for Disease Diagnosis
1 Introduction
2 Method
2.1 Subjects and Image Preprocessing
2.2 Proposed Weighted Correlation Kernel
2.3 Architecture of the Proposed Wc-Kernel Based CNN
3 Experiments
4 Conclusion
References
Robust Contextual Bandit via the Capped-2 Norm for Mobile Health Intervention
1 Introduction
2 Preliminaries
3 Robust Contextual Bandit with Capped-2 Norm
3.1 Algorithm for the Critic Updating
3.2 Algorithm for the Actor Updating
4 Experiments
4.1 Datasets
4.2 Experiments Settings
4.3 Results and Discussion
5 Conclusions and Future Directions
References
Dynamic Multi-scale CNN Forest Learning for Automatic Cervical Cancer Segmentation
Abstract
1 Introduction
2 Proposed Cluster-Based Dynamic Multi-scale Dynamic Forest
2.1 Root Node CNN Architecture
2.2 Cascaded CNNs
2.3 Proposed CNN-Based Dynamic Multi-scale Tree (DMT)
2.4 Proposed CK+1DMF Learning Framework
3 Results and Discussion
4 Conclusion
References
Multi-task Fundus Image Quality Assessment via Transfer Learning and Landmarks Detection
1 Introduction
2 Dataset
3 Method
3.1 Multi-task Convolutional Neural Networks
3.2 Optic Disc and Fovea Detection
3.3 Learning
4 Experiments and Results
5 Conclusion
References
End-to-End Lung Nodule Detection in Computed Tomography
Abstract
1 Introduction
2 Methodology
2.1 Overview
2.2 Reconstruction Sub-network
2.3 Detection Sub-network
2.4 End-to-End Fine Tuning
2.5 Inference
3 Simulation Setup
3.1 Data Source
3.2 Training Parameters
3.3 Evaluation
4 Results
4.1 FROC Analysis
4.2 Reconstructed Images
5 Conclusion and Discussion
References
CT Image Enhancement Using Stacked Generative Adversarial Networks and Transfer Learning for Lesion Segmentation Improvement
1 Introduction
2 Methods
2.1 CT Image Enhancement
2.2 Lesion Segmentation
3 Experimental Results and Analyses
4 Conclusions
References
Deep Learning Based Inter-modality Image Registration Supervised by Intra-modality Similarity
Abstract
1 Introduction
2 Method
2.1 Loss Function Based on Intra-modality Similarity
2.2 Inter-modality Registration Network
2.3 Spatial Transformation Layer
3 Experimental Results
3.1 Registration Results
4 Conclusion
References
Regional Abnormality Representation Learning in Structural MRI for AD/MCI Diagnosis
1 Introduction
2 Materials and Preprocessing
3 Proposed Method
3.1 Regional Abnormality Representation
3.2 Brain-Wise Feature Extraction and Classifier Learning
4 Experimental Settings and Results
4.1 Experimental Settings
4.2 Results and Discussion
5 Conclusion
References
Joint Registration And Segmentation Of Xray Images Using Generative Adversarial Networks
1 Introduction
2 Methods
2.1 Joint Registration and Segmentation Using GANs
2.2 Deformation Field Consistency
2.3 Obtaining Segmentation Mask
3 Experiments
3.1 Results on NIH dataset
4 Conclusion
References
SCCA-Ref: Novel Sparse Canonical Correlation Analysis with Reference to Discover Independent Spatial Associations Between White Matter Hyperintensities and Atrophy
1 Introduction
2 Method
2.1 Classical CCA for Joint Analysis of WMH and Atrophy
2.2 SCCA-Ref
2.3 Regional WMH Burden
3 Experiments
4 Conclusions
References
Synthesizing Dynamic MRI Using Long-Term Recurrent Convolutional Networks
1 Introduction
2 Materials and Methods
2.1 Network Architecture
3 Results and Discussion
References
Automatically Designing CNN Architectures for Medical Image Segmentation
1 Introduction
2 Methods
2.1 Policy Gradient
2.2 Proposed Base-Architecture for Image Segmentation
2.3 Learnable Hyperparameters
3 Experiments and Results
4 Discussions and Conclusion
References
Rotation Invariance and Directional Sensitivity: Spherical Harmonics versus Radiomics Features
1 Introduction
2 Materials and Methods
2.1 Notations
2.2 Local Rotation Invariance and Directional Sensitivity
2.3 Spherical Harmonic Wavelets
3 Results
3.1 LRI and DS of Popular Radiomics Operators
3.2 3D Synthetic Texture Classification
4 Conclusions
References
Can Dilated Convolutions Capture Ultrasound Video Dynamics?
1 Introduction
2 Materials and Methods
2.1 Data and Experimental Setup
2.2 Network Architecture
2.3 Attention Mask Module
3 Results
4 Discussion and Conclusion
References
Topological Correction of Infant Cortical Surfaces Using Anatomically Constrained U-Net
1 Introduction
2 Method
2.1 Training Set Construction
2.2 Anatomically Constrained U-Net
2.3 Inferring the New Labels of Candidate Voxels
3 Experiments
3.1 Dataset and Experimental Settings
3.2 Result
4 Conclusion
References
Self-taught Learning with Residual Sparse Autoencoders for HEp-2 Cell Staining Pattern Recognition
1 Introduction
2 Method
2.1 Material
2.2 Residual SAE for Self-taught Learning
2.3 The Aggregated Activation of the Residual SAE for Image Representation
3 Experimental Results
4 Conclusions
References
Semantic-Aware Generative Adversarial Nets for Unsupervised Domain Adaptation in Chest X-Ray Segmentation
1 Introduction
2 Method
2.1 Segmentation Network Established on Source Domain
2.2 Image Transformation with Semantic-Aware GANs
2.3 Learning Procedure and Implementation Details
3 Experimental Results
4 Conclusion
References
Brain Status Prediction with Non-negative Projective Dictionary Learning
1 Introduction
2 The Proposed Approach
2.1 Non-negative Projective Dictionary Learning
2.2 Training Algorithm
2.3 Classification
2.4 Complexity and Convergence
3 Experiments
3.1 Prediction of MCI-to-AD Conversion
3.2 Prediction of Brain Age
4 Conclusion
References
Classification of Pancreatic Cystic Neoplasms Based on Multimodality Images
1 Introduction
2 Methodology
2.1 Framework of PCN-Net
2.2 First Stage: Feature Extraction and Region Proposal
2.3 Second Stage: Intra-modality Localization and Inter-modality Registration
2.4 Third Stage: Modality Fusion and Classification
3 Experiment
3.1 Dataset and Annotation
3.2 Implementation Details
3.3 Result and Comparisons
4 Discussion and Conclusion
References
Retinal Blood Vessel Segmentation Using a Fully Convolutional Network - Transfer Learning from Patch- to Image-Level
1 Introduction
2 Proposed Model
2.1 Fully Convolutional Networks and Transfer Learning
2.2 The Proposed Framework
2.3 Neural Network Architecture
3 Experiments
3.1 Data Preparation and Network Training
3.2 Results
3.3 Discussion
4 Summary and Perspectives
References
Combining Deep Learning and Active Contours Opens The Way to Robust, Automated Analysis of Brain Cytoarchitectonics
1 Introduction
2 Methodology
2.1 Sample Preparation
2.2 Image Data
2.3 Cell Segmentation Workflow
3 Results
4 Conclusions
References
Latent3DU-net: Multi-level Latent Shape Space Constrained 3D U-net for Automatic Segmentation of the Proximal Femur from Radial MRI of the Hip
1 Introduction
2 Methods
2.1 Spatial Transform
2.2 Segmentation of the Proximal Femur
3 Experiments and Results
3.1 Dataset and Preprocessing
3.2 Training
3.3 Testing and Evaluation
3.4 Results
4 Conclusions
References
Adversarial Image Registration with Application for MR and TRUS Image Fusion
1 Introduction
2 Adversarial Image Registration (AIR)
2.1 Generator and Discriminator Networks
2.2 Adversarial Training
3 Experiments
3.1 Materials and Training
3.2 Experimental Results
4 Conclusions
References
Reproducible White Matter Tract Segmentation Using 3D U-Net on a Large-scale DTI Dataset
1 Introduction
2 Methods
2.1 Model
2.2 Dataset
2.3 Preprocessing
2.4 Reference Standard
2.5 Evaluation Metrics
3 Experiments
3.1 Method Optimization Experiments
3.2 Validation Experiments
4 Results
4.1 Method Optimization Results
4.2 Validation Results
5 Discussion
References
Competition vs. Concatenation in Skip Connections of Fully Convolutional Networks
1 Introduction
2 Methodology
2.1 Local Competition - Competitive Dense Block
2.2 Global Competition - Competitive Un-pooling Block (CUB)
2.3 Competitive Dense Fully Convolutional Network- CDFNet
3 Results and Discussion
4 Conclusion
References
Ensemble of Multi-sized FCNs to Improve White Matter Lesion Segmentation
1 Introduction
2 Dice as Evaluation Metric and Objective Function: Issues
3 Remedy: Two-Stage-Multi-sized FCNs and a New Activation Function
4 Experiments
4.1 Experimental Results
5 Conclusions
References
Automatic Accurate Infant Cerebellar Tissue Segmentation with Densely Connected Convolutional Network
Abstract
1 Introduction
2 Method
2.1 Dataset and Preprocessing
2.2 Network Architecture Design
2.3 Network Training
3 Experimental Results
3.1 Performance on 12-Month-Old Subjects
3.2 Performance on 6-Month-Old Infants
4 Conclusions
Acknowledgements
References
Nuclei Detection Using Mixture Density Networks
1 Introduction
2 Mixture Density Networks
2.1 Extending MDN for Nuclei Detection
3 Experimental Results
4 Conclusion
References
Attention-Guided Curriculum Learning for Weakly Supervised Classification and Localization of Thoracic Diseases on Chest Radiographs
1 Introduction
2 Method
2.1 A CNN Based Classification and Localization Framework
2.2 Disease Severity-Level Based Curriculum Learning
2.3 Attention Guided Iterative Refinement
3 Experiments
4 Conclusion
References
Graph of Hippocampal Subfields Grading for Alzheimer's Disease Prediction
1 Introduction
2 Materials and Methods
3 Results and Discussions
4 Conclusions
References
Deep Multiscale Convolutional Feature Learning for Weakly Supervised Localization of Chest Pathologies in X-ray Images
1 Introduction
2 Methodology
2.1 Classification-CNN
2.2 Class Aware Training of Convolutional Features
2.3 Pathology Localization by Attention CNN (A-CNN)
3 Experiments
4 Conclusion
References
Combining Heterogeneously Labeled Datasets For Training Segmentation Networks
1 Introduction
2 Methods
2.1 Naive Masking
2.2 Super Label Aware Crossentropy Loss
3 Experiments and Results
3.1 Data
3.2 Network Architecture and Training
3.3 Evaluation
3.4 Discussion and Conclusion
References
SoLiD: Segmentation of Clostridioides Difficile Cells in the Presence of Inhomogeneous Illumination Using a Deep Adversarial Network
1 Introduction
2 Methods
3 Experimental Results
4 Conclusion
References
On the Adaptability of Unsupervised CNN-Based Deformable Image Registration to Unseen Image Domains
1 Introduction
2 Materials and Methods
2.1 Datasets and Clinical Context
2.2 Unsupervised CNN-Based Image Registration
2.3 Fine-Tuning and One-Shot Learning in the Context of Unsupervised CNN-Based Image Registration
3 Results and Discussion
4 Conclusions and Future Works
References
Early Diagnosis of Autism Disease by Multi-channel CNNs
Abstract
1 Introduction
2 Materials and Methods
3 Experiments and Results
4 Conclusion
Acknowledgments
References
Longitudinal and Multi-modal Data Learning via Joint Embedding and Sparse Regression for Parkinson's Disease Diagnosis
Abstract
1 Introduction
2 Related Work
3 Methodology
3.1 System Overview
3.2 Notations
3.3 Proposed Method
4 Experiment
4.1 Image Preprocessing
4.2 Experimental Setting
4.3 Classification Performance
4.4 Regression Performance
5 Conclusion
References
Prostate Cancer Classification on VERDICT DW-MRI Using Convolutional Neural Networks
1 Introduction
2 Methods
2.1 VERDICT DW-MRI Data
2.2 Fully Convolutional Neural Networks
3 Results
4 Conclusion
References
Detection of the Pharyngeal Phase in the Videofluoroscopic Swallowing Study Using Inflated 3D Convolutional Networks
1 Introduction
2 Methodology
2.1 Dataset
2.2 Generating Pharyngeal Phase Candidates Using Optical Flow
2.3 Training Inflated 3D Convolutional Networks Using RGB/Optical Flow/Joint
3 Experimental Results
4 Conclusion
References
End-To-End Alzheimer's Disease Diagnosis and Biomarker Identification
1 Introduction
2 Dataset and Preprocessing
3 3D-CNN Training and Evaluation
4 Experiment Results
5 Conclusion
References
Small Organ Segmentation in Whole-Body MRI Using a Two-Stage FCN and Weighting Schemes
1 Introduction
2 Materials and Methods
2.1 Materials
2.2 Two-Stage Network: A Coarse-to-Fine Approach
3 Experiments and Results
4 Discussion and Conclusion
References
Masseter Segmentation from Computed Tomography Using Feature-Enhanced Nested Residual Neural Network
1 Introduction
2 Methods
2.1 Masseter Region Location
2.2 Feature-Enhanced Nested Residual Neural Network
3 Experiments
4 Discussion and Conclusion
References
Iterative Interaction Training for Segmentation Editing Networks
1 Introduction
2 Methods
3 Experiments and Results
4 Conclusions
References
Temporal Consistent 2D-3D Registration of Lateral Cephalograms and Cone-Beam Computed Tomography Images
1 Introduction
2 Method
2.1 CNN-Based 2D-3D Registration
2.2 Temporal Consistent 2D-3D Registration
3 Experiments
4 Conclusion
References
Computation of Total Kidney Volume from CT Images in Autosomal Dominant Polycystic Kidney Disease Using Multi-task 3D Convolutional Neural Networks
Abstract
1 Introduction
2 Materials and Methods
2.1 Dataset and Preprocessing
2.2 Multi-task 3D Fully Convolutional Network for ADPK Segmentation
2.3 Bootstrapping Cross Entropy Loss
3 Experiments and Training
4 Results
5 Conclusion and Future Work
Acknowledgements
References
Dynamic Routing on Deep Neural Network for Thoracic Disease Classification and Sensitive Area Localization
1 Introduction
2 Methods
2.1 1 1 Convolutional Capsule Layer
3 Experiment Results
4 Conclusion
References
Deep Learning for Fast and Spatially-Constrained Tissue Quantification from Highly-Undersampled Data in Magnetic Resonance Fingerprinting (MRF)
Abstract
1 Introduction
2 Materials and Method
2.1 Data Acquisition and Pre-processing
2.2 Proposed U-Net Model
3 Experiments
3.1 Experimental Settings
3.2 Results
4 Conclusion
References
Correction to: Deep Learning for Fast and Spatially-Constrained Tissue Quantification from Highly-Undersampled Data in Magnetic Resonance Fingerprinting (MRF)
Correction to: Chapter "Deep Learning for Fast and Spatially-Constrained Tissue Quantification from Highly-Undersampled Data in Magnetic Resonance Fingerprinting (MRF)" in: Y. Shi et al. (Eds.): Machine Learning in Medical Imaging, LNCS 11046, https://doi.org/10.1007/978-3-030-00919-9_46
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Machine Learning in Medical Imaging

Description

More details

Other editions

Additional editions

Content

System requirements