Understanding and Interpreting Machine Learning in Medical Image Computing Applications
First International Workshops, MLCN 2018, DLF 2018, and iMIMIC 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16-20, 2018, Proceedings
Published on 23. October 2018
XVI, 149 pages
978-3-030-02628-8 (ISBN)
Description
This book constitutes the refereed joint proceedings of the First International Workshop on Machine Learning in Clinical Neuroimaging, MLCN 2018, the First International Workshop on Deep Learning Fails, DLF 2018, and the First International Workshop on Interpretability of Machine Intelligence in Medical Image Computing, iMIMIC 2018, held in conjunction with the 21st International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2018, in Granada, Spain, in September 2018.
The 4 full MLCN papers, the 6 full DLF papers, and the 6 full iMIMIC papers included in this volume were carefully reviewed and selected. The MLCN contributions develop state-of-the-art machine learning methods such as spatio-temporal Gaussian process analysis, stochastic variational inference, and deep learning for applications in Alzheimer's disease diagnosis and multi-site neuroimaging data analysis; the DLF papers evaluate the strengths and weaknesses of DL and identifythe main challenges in the current state of the art and future directions; the iMIMIC papers cover a large range of topics in the field of interpretability of machine learning in the context of medical image analysis.
More details
Other editions
Additional editions
Danail Stoyanov | Zeike Taylor | Seyed Mostafa Kia
Understanding and Interpreting Machine Learning in Medical Image Computing Applications
First International Workshops, MLCN 2018, DLF 2018, and iMIMIC 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16-20, 2018, Proceedings
Book
10/2018
Springer
€53.49
Shipment within 7-9 days
Content
- Intro
- Additional Workshop Editors
- MLCN 2018 Preface
- DLF 2018 Preface
- iMIMIC 2018 Preface
- Organization
- Contents
- First International Workshop on Machine Learning in Clinical Neuroimaging, MLCN 2018
- Alzheimer's Disease Modelling and Staging Through Independent Gaussian Process Analysis of Spatio-Temporal Brain Changes
- 1 Introduction
- 2 Method
- 3 Results
- 3.1 Benchmark on Synthetic Data
- 3.2 Application on Real Data
- 4 Conclusion
- References
- Multi-channel Stochastic Variational Inference for the Joint Analysis of Heterogeneous Biomedical Data in Alzheimer's Disease
- 1 Introduction
- 2 Method
- 2.1 Multi-channel Variational Inference
- 2.2 Gaussian Linear Case
- 3 Experiments
- 3.1 Experiments on Linearly Generated Synthetic Datasets
- 3.2 Application to Clinical and Medical Imaging Data in AD
- 4 Discussion and Conclusion
- References
- Visualizing Convolutional Networks for MRI-Based Diagnosis of Alzheimer's Disease
- 1 Introduction
- 2 Related Work
- 2.1 Alzheimer Classification
- 2.2 Visualization Methods
- 3 Methods
- 3.1 Data
- 3.2 Model
- 3.3 Visualization Methods
- 4 Results
- 4.1 Classification
- 4.2 Relevant Brain Areas
- 4.3 Differences Between Visualization Methods
- 5 Conclusion
- References
- Finding Effective Ways to (Machine) Learn fMRI-Based Classifiers from Multi-site Data
- 1 Introduction
- 1.1 Multi-site Data and Batch Effects
- 2 Machine Learning and Functional Connectivity Graphs
- 3 Batch Effects Correction Techniques
- 3.1 Adding Site as Covariate
- 3.2 Z-Score Normalization
- 3.3 Whitening
- 3.4 Solving Linear Transformations
- 4 Experiments and Results
- 4.1 Dataset
- 4.2 Experiments and Results
- 5 Discussion
- References
- First International Workshop on Deep Learning Fails Workshop, DLF 2018
- Towards Robust CT-Ultrasound Registration Using Deep Learning Methods
- 1 Introduction
- 2 Methods
- 3 Data
- 3.1 Clinical Data
- 3.2 Training Data
- 4 Experiments
- 4.1 Mono-Modal
- 4.2 Multi-modal (Simulated)
- 4.3 Inaccurate Ground Truth
- 4.4 CT-US
- 5 Discussion and Conclusion
- References
- To Learn or Not to Learn Features for Deformable Registration?
- 1 Introduction
- 2 Method
- 2.1 Discrete Optimization
- 2.2 Deep Learning Framework
- 3 Experiments and Results
- 3.1 Datasets Description
- 3.2 Evaluation Metric
- 3.3 Implementation Detail
- 3.4 Feature Learning Experiments and Results
- 4 Conclusions
- References
- Evaluation of Strategies for PET Motion Correction - Manifold Learning vs. Deep Learning
- 1 Introduction
- 2 Methods
- 2.1 Network Architecture
- 2.2 Training Details
- 3 Experiments
- 3.1 Synthetic Dataset
- 3.2 Comparison Method: Data-Driven Gating
- 3.3 Assessment of Corrected Volume Quality
- 4 Discussion and Conclusions
- References
- Exploring Adversarial Examples
- 1 Introduction
- 2 Related Work
- 3 Methods
- 3.1 Dataset
- 3.2 Training
- 3.3 Adversarial Data Creation
- 4 Results
- 5 Discussion and Conclusion
- References
- Shortcomings of Ventricle Segmentation Using Deep Convolutional Networks
- 1 Introduction
- 2 Methods
- 2.1 Data and Preprocessing
- 2.2 Ventricle Segmentation Network
- 2.3 Training Procedure
- 3 Experiments and Results
- 4 Discussion and Conclusions
- References
- Vulnerability Analysis of Chest X-Ray Image Classification Against Adversarial Attacks
- 1 Introduction
- 2 Methods
- 2.1 Applied Deep Networks
- 2.2 Applied Adversarial Attacks
- 3 Dataset
- 4 Results and Discussion
- 5 Conclusion
- References
- First International Workshop on Interpretability of Machine Intelligence in Medical Image Computing, iMIMIC 2018
- Collaborative Human-AI (CHAI): Evidence-Based Interpretable Melanoma Classification in Dermoscopic Images
- 1 Introduction
- 2 Methods
- 2.1 Triplet-Loss with Global Average Pooling
- 2.2 Hierarchical Triplet Selection Logic
- 2.3 Experimental Design
- 3 Results
- 4 Conclusion
- References
- Automatic Brain Tumor Grading from MRI Data Using Convolutional Neural Networks and Quality Assessment
- 1 Introduction
- 2 Methods
- 2.1 Extraction of the Region of Interest
- 2.2 Glioma Grading CNN
- 2.3 Grade Prediction Interpretability
- 3 Experimental Setup
- 4 Results and Discussion
- 5 Conclusion
- References
- Visualizing Convolutional Neural Networks to Improve Decision Support for Skin Lesion Classification
- 1 Introduction
- 2 Related Work
- 3 Architecture and Training
- 4 Feature Map Visualization
- 5 Conclusion
- References
- Regression Concept Vectors for Bidirectional Explanations in Histopathology
- 1 Introduction
- 2 Methods
- 2.1 Correlation to Network Prediction
- 2.2 Regression Concept Vectors
- 2.3 Sensitivity to RCV
- 2.4 Evaluation of the Explanations
- 3 Experiments and Results
- 3.1 Datasets
- 3.2 Network Architecture and Training
- 3.3 Results
- 4 Discussion and Future Work
- References
- Towards Complementary Explanations Using Deep Neural Networks
- 1 Introduction
- 1.1 Satisfying the Curiosity of Decision Makers
- 2 Complementary Explanations Using Deep Neural Networks
- 3 The Three Cs of Interpretability
- 4 Experimental Assessment
- 5 Conclusion
- References
- How Users Perceive Content-Based Image Retrieval for Identifying Skin Images
- 1 Introduction
- 2 Method
- 2.1 Study Design
- 2.2 Dataset
- 2.3 System Description
- 2.4 Protocol
- 2.5 Data Collection
- 3 Results
- 4 Discussion
- References
- Author Index
