Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support

Name: Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support | Third International Workshop, DLMIA 2017, and 7th International Workshop, ML-CDS 2017, Held in Conjunction with MICCAI 2017, Québec City, QC, Canada, September 14, Proceedings
Brand: Springer
Price: 69.54 EUR
Availability: OnlineOnly

Third International Workshop, DLMIA 2017, and 7th International Workshop, ML-CDS 2017, Held in Conjunction with MICCAI 2017, Québec City, QC, Canada, September 14, Proceedings

M. Jorge Cardoso Tal Arbel Gustavo Carneiro Tanveer Syeda-Mahmood João Manuel R.S. Tavares Mehdi Moradi Andrew Bradley Hayit Greenspan João Paulo Papa Anant Madabhushi Jacinto C. Nascimento Jaime S. Cardoso Vasileios Belagiannis Zhi Lu(Editor)

Springer (Publisher)

Published on 7. September 2017

XIX, 385 pages

E-Book

PDF with digital watermarking

System requirements

978-3-319-67558-9 (ISBN)

€69.54incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Workshop Editors
Preface DLMIA 2017
Organization
Preface ML-CDS 2017
Organization
Contents
Third International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017
Simultaneous Multiple Surface Segmentation Using Deep Learning
1 Introduction
2 Method
3 Experiments
4 Results
5 Discussion and Conclusion
References
A Deep Residual Inception Network for HEp-2 Cell Classification
Abstract
1 Introduction
2 Deep Residual Inception
2.1 Network Architecture
2.2 DRI Module
2.3 Network Training
3 Results
3.1 Dataset
3.2 Data Augmentation
3.3 Performance Analysis
3.4 Comparisons
4 Conclusion
References
Joint Segmentation of Multiple Thoracic Organs in CT Images with Two Collaborative Deep Architectures
1 Introduction
2 Method
2.1 SharpMask Feature Fusion Architecture and CRF Refinement
2.2 Learning Anatomical Constraints
3 Experiments
3.1 Dataset and Pre-processing
3.2 Training
3.3 Results
4 Conclusions
References
Accelerated Magnetic Resonance Imaging by Adversarial Neural Network
1 Introduction
2 Method
2.1 K-space
2.2 Objective
2.3 Network Architecture
3 Experimental Results
4 Conclusions
References
Left Atrium Segmentation in CT Volumes with Fully Convolutional Networks
1 Introduction
2 Related Work
3 Method
3.1 Preprocessing
3.2 Fully Convolutional Network
3.3 Shape Constraints
4 Experiments
5 Conclusion
References
3D Randomized Connection Network with Graph-Based Inference
1 Introduction
2 Methodology
2.1 Convolutional LSTM and 3D Convolution
2.2 Randomized Connection Network
2.3 Graph-Based Inference
3 Experiment
4 Conclusion
References
Adversarial Training and Dilated Convolutions for Brain MRI Segmentation
1 Introduction
2 Materials and Methods
2.1 Data
2.2 Network Architecture
2.3 Adversarial Training
3 Experiments and Results
3.1 Experiments
3.2 Evaluation
4 Discussion and Conclusions
References
CNNs Enable Accurate and Fast Segmentation of Drusen in Optical Coherence Tomography
1 Introduction
2 Related Work
3 Method
3.1 Data Preparation
3.2 Network Architecture and Training
3.3 Three Approaches to Drusen Segmentation
4 Experiments and Results
4.1 Cross-Validation Setup
4.2 Quantitative Evaluation
4.3 Robustness to Additional Pathology
4.4 3D Visualization of Results
5 Conclusion
References
Region-Aware Deep Localization Framework for Cervical Vertebrae in X-Ray Images
1 Introduction
2 Data
3 Methodology
3.1 Localization Ground Truth
3.2 Network Architectures
3.3 Training
3.4 Region-Aware Term
3.5 Updated Loss Function
3.6 Experiments and Inference
4 Results and Discussions
5 Conclusion
References
Domain-Adversarial Neural Networks to Address the Appearance Variability of Histopathology Images
1 Introduction
2 Materials and Methods
2.1 Datasets
2.2 The Underlying CNN Architecture
2.3 Three Approaches to Handling Appearance Variability
2.4 Evaluation
3 Experiments and Results
4 Discussion and Conclusions
References
Accurate Lung Segmentation via Network-Wise Training of Convolutional Networks
1 Introduction
2 Methods
2.1 Lung Segmentation with Atrous Convolutions
2.2 Network-Wise Training of CNN
3 Computational Experiments
3.1 Performance Metrics
3.2 Quantatitive and Qualititive Results
4 Conclusion
References
Deep Residual Recurrent Neural Networks for Characterisation of Cardiac Cycle Phase from Echocardiograms
1 Introduction
2 Methods
2.1 Dataset
2.2 Deep Residual Recurrent Neural Networks (RRNs)
3 Experiments
4 Results and Discussion
5 Conclusion and Future Works
References
Computationally Efficient Cardiac Views Projection Using 3D Convolutional Neural Networks
1 Introduction
2 Methods
3 Results
4 Conclusion
References
Non-rigid Craniofacial 2D-3D Registration Using CNN-Based Regression
1 Introduction
2 Methods
2.1 Regression-Based 2D-3D Registration
2.2 CNN Based Regressor
3 Experiments
3.1 Qualitative Assessment
4 Conclusion
References
A Deep Level Set Method for Image Segmentation
1 Introduction
2 Methods
2.1 The Level Set Method
2.2 The Integrated FCN-Levelset Model
3 Experiments and Results
3.1 Data
3.2 Experiments
3.3 Results
4 Discussion
References
Context-Based Normalization of Histological Stains Using Deep Convolutional Features
1 Introduction
2 Method
2.1 Feature-Aware Normalization
2.2 Normalization by Denoising
3 Experiments
4 Discussion
References
Transitioning Between Convolutional and Fully Connected Layers in Neural Networks
1 Introduction
2 Related Work
3 Method
3.1 Inception Module
3.2 Transition Module
4 Experiment
5 Results
5.1 Experiment 1: Comparison with Regularizers
5.2 Experiment 2: Comparing Architectures
5.3 Experiment 3: BreaKHis
6 Conclusion
References
Quantifying the Impact of Type 2 Diabetes on Brain Perfusion Using Deep Neural Networks
1 Introduction
2 Materials
3 Methods
3.1 Compute Mean Gray Matter CBF per Anatomical Region
3.2 Identify candidate regions for further analysis
3.3 Estimate Candidate Region Association Using a DNN
4 Results
4.1 Performance Comparison of the Learning Models
4.2 Statistical Significance of the Proposed Model
5 Discussion
6 Conclusions
References
Multi-stage Diagnosis of Alzheimer's Disease with Incomplete Multimodal Data via Multi-task Deep Learning
1 Introduction
2 Method
2.1 Multi-task Learning
2.2 Multi-task Deep Learning for Incomplete Multimodal Data
3 Materials, Preprocessing and Feature Extraction
4 Results and Discussions
5 Conclusion
References
A Multi-scale CNN and Curriculum Learning Strategy for Mammogram Classification
1 Introduction
2 Multi-scale CNN with Curriculum Learning Strategy
3 Experiments
4 Conclusions
References
Analyzing Microscopic Images of Peripheral Blood Smear Using Deep Learning
1 Introduction
2 The Shonit?System for Analysis of Peripheral Blood Smears
3 Deep Learning Techniques for Analyzing PBS Images
3.1 Cell Extraction
3.2 Cell Classification
4 Experimental Results
5 Conclusion
References
AGNet: Attention-Guided Network for Surgical Tool Presence Detection
1 Introduction
2 Attention-Guided Network
2.1 Global Prediction Network
2.2 Local Prediction Network
3 Experiments
3.1 Datasets and Preprocessing
3.2 Training Procedure
3.3 Ablation Analysis
3.4 Comparison with the State-of-the-Arts
4 Conclusion
References
Pathological Pulmonary Lobe Segmentation from CT Images Using Progressive Holistically Nested Neural Networks and Random Walker
1 Introduction
2 Method
2.1 Lobar Boundary Segmentation
2.2 3D Random Walker
3 Experiments and Results
4 Conclusion
References
End-to-End Unsupervised Deformable Image Registration with a Convolutional Neural Network
1 Introduction
2 Method
3 Data
4 Experiments and Results
4.1 Registration of Handwritten Digits
4.2 Registration of Cardiac MRI
5 Discussion and Conclusion
References
Stain Colour Normalisation to Improve Mitosis Detection on Breast Histology Images
1 Introduction
2 Related Work
3 Dataset
4 Method
4.1 Patch Generation
4.2 CNN Architecture
4.3 Training and Testing Workflow
5 Results and Discussion
6 Conclusion
References
3D FCN Feature Driven Regression Forest-Based Pancreas Localization and Segmentation
1 Introduction
2 Method
2.1 Overview
2.2 Pancreas Localization
2.3 Patient-Specific Probabilistic Atlas Generation and Pancreas Segmentation
3 Experiments and Discussion
References
A Unified Framework for Tumor Proliferation Score Prediction in Breast Histopathology
1 Introduction
2 Methodology
2.1 Whole Slide Image Handling
2.2 Deep Convolutional Neural Networks Based Mitosis Detection
2.3 Tumor Proliferation Score Prediction
3 Results
3.1 Datasets
3.2 Experiments
4 Conclusion
References
Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations
1 Introduction
2 Methods
2.1 Loss Functions for Unbalanced Data
2.2 Deep Learning Framework
3 Experiments and Results
3.1 Experiments
3.2 2D Results
3.3 3D Results
4 Discussion
References
ssEMnet: Serial-Section Electron Microscopy Image Registration Using a Spatial Transformer Network with Learned Features
1 Introduction
2 Method
2.1 Feature Generation Using a Convolutional Autoencoder
2.2 Deformable Image Registration Using a Spatial Transformer Network
3 Results
4 Discussion and Conclusion
References
Fully Convolutional Regression Network for Accurate Detection of Measurement Points
1 Introduction
2 Related Work
3 Regressing Point Locations
3.1 Fully Convolutional Network with Center of Mass Layer
3.2 Convolutional Long Short-Term Memory for Temporal Consistency
4 Results
5 Conclusion
References
Fast Predictive Simple Geodesic Regression
1 Introduction
2 Fast Predictive Simple Geodesic Regression (FPSGR)
3 Experimental Setup and Training of Prediction Models
4 Discussion of Experimental Results for 3D ADNI Data
5 Conclusion and Future Work
References
Learning Spatio-Temporal Aggregation for Fetal Heart Analysis in Ultrasound Video
1 Introduction
1.1 Related Work
2 Fetal Heart Description with Spatio-Temporal CNNs
2.1 Spatial Baseline Models
2.2 Direct Temporal Encoding
2.3 Hierarchical Temporal Encoding
3 Implementation Details
4 Experimental Results and Comparisons
5 Conclusions
References
Fast, Simple Calcium Imaging Segmentation with Fully Convolutional Networks
1 Introduction
2 Related Work
3 Data and Metrics
3.1 Calcium Imaging Datasets
3.2 Summary Images
3.3 Evaluation Metrics
4 Proposed Models
4.1 U-Net with 2D Summary Images
4.2 Other Deep Learning Models
5 Conclusion
References
Self-supervised Learning for Spinal MRIs
1 Introduction
2 Self-supervised Learning
2.1 Extracting Vertebral Bodies and Intervertebral Discs
2.2 Longitudinal Learning via Contrastive Loss
2.3 Auxiliary Loss -- Predicting VB Levels
2.4 Architecture
3 Dataset and Implementation Details
3.1 Radiological Gradings
3.2 Training
4 Experiments and Results
4.1 Longitudinal Learning and Self-supervision
4.2 Benefits of Pre-training
5 Conclusion
References
Skin Lesion Segmentation via Deep RefineNet
Abstract
1 Introduction
2 Methodology
2.1 Residual Learning
2.2 Deep RefineNet
2.3 Conditional Random Field for Boundary Refinement
3 Experiments
3.1 Dataset and Implementation
3.2 Results
4 Conclusion
References
Multi-scale Networks for Segmentation of Brain Magnetic Resonance Images
Abstract
1 Introduction
2 Algorithm
2.1 Multi-scale U-Nets for Voxel Classification
2.2 Multi-scale BPNNs for Voxel Classification
2.3 CAE-Based Image Restoration
2.4 Implementation
3 Experiments and Results
4 Discussions
5 Conclusion
Acknowledgement
References
Deep Learning for Automatic Detection of Abnormal Findings in Breast Mammography
Abstract
1 Introduction
2 Methods
3 Experiments and Results
4 Summary
References
Grey Matter Segmentation in Spinal Cord MRIs via 3D Convolutional Encoder Networks with Shortcut Connections
1 Introduction
2 Methods
2.1 Network
2.2 Training
2.3 Segmentation
3 Experiments and Results
4 Conclusions
References
7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017
Mapping Multi-Modal Routine Imaging Data to a Single Reference via Multiple Templates
1 Introduction
2 Method
2.1 Fragment to Template Registration
2.2 Template to Reference Registration
2.3 Template Selection
3 Experiments
4 Results
5 Conclusion
References
Automated Detection of Epileptogenic Cortical Malformations Using Multimodal MRI
Abstract
1 Introduction
2 Methods
2.1 MRI Acquisition
2.2 Multi-contrast MRI Pre-processing
2.3 Multi-surface Generation
2.4 Feature Extraction
3 Experiment and Results
3.1 Subjects
3.2 Manual Lesion Segmentation
3.3 Classification Paradigm (Fig. 1)
3.4 Results
4 Discussion
References
Prediction of Amyloidosis from Neuropsychological and MRI Data for Cost Effective Inclusion of Pre-symptomatic Subjects in Clinical Trials
1 Introduction
2 Materials and Methods
2.1 Validation Cohorts
2.2 Input Features
2.3 Algorithms
2.4 Performance Measures
3 Experiments and Results
3.1 Algorithm Benchmark
3.2 Feature Selection
3.3 Use of MRI
3.4 Generalization on an Independent Cohort
3.5 Pooling Data Sets
3.6 Effect of Sample Size
4 Conclusion
References
Automated Multimodal Breast CAD Based on Registration of MRI and Two View Mammography
1 Introduction
2 Methods
2.1 Image Registration
2.2 CAD System
2.3 Clinical Datasets and Evaluation Methods
3 Results
4 Discussion and Conclusion
References
EMR-Radiological Phenotypes in Diseases of the Optic Nerve and Their Association with Visual Function
Abstract
1 Introduction
2 Methods
2.1 Data
2.2 Outcomes: Visual Function Scores
2.3 Image Processing
2.4 EMR Features
2.5 Dimensionality Reduction: PCA and MCA
2.6 Stepwise Generalized Linear Model
2.7 Test of Deviance
3 Results
4 Discussion
Acknowledgements
References
Erratum to: Fast Predictive Simple Geodesic Regression
Erratum to: Chapter "Fast Predictive Simple Geodesic Regression" in: M.J. Cardoso et al. (Eds.), Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, LNCS 10553, https://doi.org/10.1007/978-3-319-67558-9_31
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support

Description

More details

Other editions

Additional editions

Content

System requirements