Medical Image Understanding and Analysis

Name: Medical Image Understanding and Analysis | 26th Annual Conference, MIUA 2022, Cambridge, UK, July 27-29, 2022, Proceedings
Brand: Springer
Price: 106.99 EUR
Availability: OnlineOnly

26th Annual Conference, MIUA 2022, Cambridge, UK, July 27-29, 2022, Proceedings

Guang Yang Angelica Aviles-Rivero Michael Roberts Carola-Bibiane Schönlieb(Editor)

Springer (Publisher)

Published on 25. July 2022

XVII, 911 pages

E-Book

PDF with digital watermarking

System requirements

978-3-031-12053-4 (ISBN)

€106.99incl. 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
Biomarker Detection
Neck Fat Estimation from DXA Using Convolutional Neural Networks
1 Introduction
2 Methods
3 Results
4 Discussion and Conclusion
References
Multimodal Cardiomegaly Classification with Image-Derived Digital Biomarkers
1 Introduction
2 Data and Methods
2.1 Data Sources
2.2 Dataset Preparation
2.3 Pre-processing
2.4 Models
2.5 CTR and CPAR: Computation and Model Integration
3 Results
3.1 CTR Computation
3.2 Multimodal Classification
4 Discussion
4.1 Principal Findings
4.2 Comparison to Related Works
4.3 Strengths and Weaknesses of the Study
References
Proton Density Fat Fraction of Breast Adipose Tissue: Comparison of the Effect of Fat Spectra and Initial Evaluation as a Biomarker
1 Introduction
2 Methods
2.1 Data Acquisition
2.2 Spectroscopic Analysis
2.3 PDFF Map Generation
2.4 Body Masking
2.5 Segmentation of Adipose Tissue from Fibrous Tissue
3 Results
3.1 Variation in PDFF with Different Fat Spectra
3.2 Variation in PDFF Calculation Between 2D and 3D CSE Acquisitions
3.3 Variation in PDFF Between Left and Right Breast
3.4 Variation in PDFF of Perilesional Adipose Tissue
4 Discussion
References
Revisiting the Shape-Bias of Deep Learning for Dermoscopic Skin Lesion Classification
1 Introduction
2 Definition and Isolation of Texture, Shape, and Color in Skin Lesions
2.1 Feature Ablations
3 Datasets and Methodology
3.1 Datasets
3.2 Experimental Setup
4 Deep Feature Extractors for Dermoscopy Can Encode Disentangled Features
4.1 Different Dermoscopic Skin Lesion Datasets Have Different Biases
4.2 Dermoscopic Skin Lesion Classifiers Learn Entangled Features
4.3 Feature Extractors Are only Partially Feature Biased
5 Dermoscopy Relies on Complex Feature Combinations in Spectral Domain
5.1 Dermoscopy Features Are Spread over Phase and Amplitude
5.2 Focusing on Amplitude Can Improve Performance
6 Discussion
7 Conclusion
References
Novel Imaging, Image Registration and Reconstruction
Joint Group-Wise Motion Estimation and Segmentation of Cardiac Cine MR Images Using Recurrent U-Net
1 Introduction
2 Methods
2.1 Group-Wise Registration for Cardiac Motion Estimation
2.2 Joint Cardiac Motion Estimation and Segmentation
2.3 Network Training with Enhancement Mask
3 Experiments and Results
3.1 Dataset Preprocessing
3.2 Evaluation
3.3 Results
4 Conclusion
References
Recursive Deformable Image Registration Network with Mutual Attention
1 Introduction
2 Methods
2.1 Image Registration
2.2 Recursive Registration Networks
2.3 Mutual Attention
3 Experiments
3.1 Datasets
3.2 Training Details
3.3 Implementation and Evaluation
4 Results
5 Discussion and Conclusion
References
Spatiotemporal Attention Constrained Deep Learning Framework for Dual-Tracer PET Imaging
1 Introduction
2 Methodology
2.1 Dual-Tracer PET Imaging Model
2.2 Network Architecture
2.3 Loss Function and Evaluation Metrics
3 Experiments and Results
3.1 Simulation Datasets and Implementation Details
3.2 Simulation Experiments
3.3 Real Experiments
4 Discussion
5 Conclusion
References
Faster Diffusion Cardiac MRI with Deep Learning-Based Breath Hold Reduction
1 Introduction
2 Background
2.1 DT-CMR
2.2 De-noising in DT-CMR
3 Methods
3.1 Data Acquisition
3.2 Data Preparation
3.3 The Effect of Repetitions
3.4 Deep-Learning-Based De-noising
3.5 DT-CMR Post-processing
3.6 DT-CMR Maps Comparison
4 Results
4.1 The Effect of Repetitions
4.2 Deep-Learning-Based De-noising
5 Discussion
5.1 Breath-Hold Choice
5.2 Deep-Learning-Based De-noising
6 Conclusion
References
Preoperative CT and Intraoperative CBCT Image Registration and Evaluation in Robotic Cochlear Implant Surgery
1 Introduction
2 Methods
2.1 CT-CBCT Registration
2.2 Evaluation
3 Experiments
3.1 Datasets
3.2 Experimental Setup
3.3 Registration Parameters
4 Results
5 Conclusion
References
Simultaneous Semantic and Instance Segmentation for Colon Nuclei Identification and Counting
1 Introduction
2 Proposed Framework
2.1 Every Single Nucleus Matters: Instance Segmentation Model
2.2 Denser Predictions for Better Performance: Semantic Segmentation Model
2.3 Model Ensemble: A Non-Maximum Suppression Embedding Algorithm
3 Experimental Results
3.1 Results and Discussion
4 Conclusion
References
Point2Mask: A Weakly Supervised Approach for Cell Segmentation Using Point Annotation
1 Introduction
2 Related Work
3 Point2Mask: The Proposed Approach
3.1 Backbone Network for Feature Extraction
3.2 Region Proposal Network for Cell Region Detection and Groundtruth Matching
3.3 Prediction Head
4 Dataset
5 Evaluation Metrics
6 Experimental Setup
6.1 Point2Mask vs Fully Supervised Method and Impact of Validated Annotated Points
6.2 Impact of Validated Annotated Points on Different Cell Cultures
7 Analysis and Discussion
8 Conclusion
References
Image Interpretation
Class Distance Weighted Cross-Entropy Loss for Ulcerative Colitis Severity Estimation
1 Introduction
2 Related Work
3 Class Distance Weighted Cross-Entropy
3.1 Motivation
3.2 Class Distance Weighted Cross-Entropy Loss Function
4 Experiments
4.1 Dataset
4.2 Training Details
4.3 Evaluation Metrics
4.4 Penalization Factor Analysis
5 Class Activation Maps (CAM)
6 Results and Discussion
7 Conclusion
References
Procrustes Analysis of Muscle Fascicle Shapes Based on DTI Fibre Tracking
1 Introduction
1.1 DTI and Fibre Tracking
1.2 Skeletal Muscle
1.3 Shape Analysis
2 Procrustes Shape Analysis
2.1 Shape
2.2 Landmark and Configuration
2.3 Pre-shape
2.4 Procrustes Distances
2.5 Ordinary and Generalised Procrustes Analysis
3 Procrustes Methods for Fascicle Shape Analysis
3.1 Landmark Construction for Fascicle Shapes Using Interpolation
3.2 Example of OPA
3.3 Example of GPA
3.4 Example of Procrustes Distance
4 Real Application: Mean Fascicle Shapes in Medial Gastrocnemius
4.1 Data Preparation
4.2 Results Analysis
5 Summary and Further Work
References
Weakly Supervised Captioning of Ultrasound Images
1 Introduction
2 Methods
3 Experiments
4 Results and Discussion
5 Conclusion and Future Work
5.1 Future Work
References
Computerised Methods for Monitoring Diabetic Foot Ulcers on Plantar Foot: A Feasibility Study
1 Introduction
2 Related Works
3 DFU Timeline Dataset
4 Methodology
4.1 Determining the Area of DFU
4.2 Site of the DFU
4.3 DFU Progress Prediction
5 Results and Discussion
6 Conclusion
References
CellCentroidFormer: Combining Self-attention and Convolution for Cell Detection
1 Introduction
2 Related Work
3 Method
3.1 Model Architecture
3.2 Bounding Ellipses for Cell Detection
4 Experiments
4.1 Datasets and Pre-processing
4.2 Baseline Model
4.3 Training Setup, Metrics, and Hyperparameters
4.4 Training Curves and Performance Comparison
5 Conclusion
References
GPU-Net: Lightweight U-Net with More Diverse Features
1 Introduction
2 Related Work
3 Methodology
3.1 GP-Module
3.2 Network Architecture
4 Experiments and Results
4.1 Dataset
4.2 Experimental Setup
4.3 Results and Discussion
5 Conclusion
References
Self-supervision and Multi-task Learning: Challenges in Fine-Grained COVID-19 Multi-class Classification from Chest X-rays
1 Introduction
2 Method
2.1 Baseline Model
2.2 Multi-task Learning
2.3 Self-supervised Pre-training
3 Dataset
4 Experimental Settings
5 Results and Discussion
5.1 Lack of Visual Cues in COVID-19 CXRs
5.2 (Mis)learned Representations
5.3 COVID-19 Binary Classifications
6 Conclusion
References
Image Segmentation
Ultrasonography Uterus and Fetus Segmentation with Constrained Spatial-Temporal Memory FCN
1 Introduction
2 Methodology
2.1 Overview
2.2 Task-Specific FCN
2.3 Bidirectional Convolutional LSTM
2.4 Joint Uterus and Fetus Segmentation
2.5 Learning with Spatial Constrained Loss
3 Experiments
4 Conclusion
References
Thigh and Calf Muscles Segmentation Using Ensemble of Patch-Based Deep Convolutional Neural Network on Whole-Body Water-Fat MRI
1 Introduction
2 Methodology
2.1 Model Architecture and Model Training
2.2 Model Inference and Post-processing
3 Experiments and Evaluation
3.1 Material
3.2 Experiments
3.3 Parameter Settings
3.4 Results
4 Conclusions and Discussion
References
Fitting Segmentation Networks on Varying Image Resolutions Using Splatting
1 Introduction
2 Methods
2.1 1D Toy Example
2.2 Splatting and Resampling
2.3 The Mean Space
3 Experiments and Results
3.1 The Baseline Network
3.2 Simulated Data: Brain Tumour Segmentation
3.3 Real Data: Brain Tissue Segmentation
4 Conclusions
References
Rotation-Equivariant Semantic Instance Segmentation on Biomedical Images
1 Introduction
2 Related Work
3 Methods
3.1 Loss Function
3.2 Rotation-Equivariance of Architecture
3.3 Rotation-Equivariance of Clustering
4 Datasets
4.1 Synthetic Scattered Sticks
4.2 BBBC038
5 Experimental Setup
5.1 Evaluation of Segmentation
5.2 Evaluation of Equivariance to Rotations
6 Results
7 Discussion
8 Conclusions
References
Joint Learning with Local and Global Consistency for Improved Medical Image Segmentation
1 Introduction
2 Related Work
3 Image-to-Patch w/Patch-to-Image (IPPI)
3.1 Loss Function
4 Experiments and Results
4.1 Data
4.2 Implementation Details
4.3 Performance Analysis
4.4 Visual Prediction
4.5 Ablation Experiments
5 Conclusion
References
LKAU-Net: 3D Large-Kernel Attention-Based U-Net for Automatic MRI Brain Tumor Segmentation
1 Introduction
2 Method
2.1 3D LK Attention
2.2 3D LK Attention-Based U-Net
3 Experiments
3.1 Data Acquisition
3.2 Preprocessing and Data Augmentation
3.3 Training and Optimization
3.4 Postprocessing
3.5 Evaluation Metrics
4 Results and Discussion
5 Conclusion
References
Attention-Fused CNN Model Compression with Knowledge Distillation for Brain Tumor Segmentation
1 Introduction
2 Method
2.1 Transformer Connection
2.2 Knowledge Distillation
2.3 Loss Function
2.4 Experimental Setting
3 Results
3.1 U-Net Architecture Compression
3.2 Transformer Connection Compression
3.3 Knowledge Distillation Between Models
4 Discussion
5 Conclusion
References
Lung Segmentation Using ResUnet++ Powered by Variational Auto Encoder-Based Enhancement in Chest X-ray Images
1 Introduction
2 Materials and Methods
2.1 Dataset
2.2 Pre-processing
2.3 ResUNet++ Architecture
2.4 Proposed Image Enhancement Approach VAEU-Net
2.5 Loss Functions
3 Experimental Results
3.1 Lung Segmentation Training
3.2 The Training of VAEU-Net
3.3 Testing of VAEU-Net and ResUnet++
4 Discussion
5 Conclusion
References
A Neural Architecture Search Based Framework for Segmentation of Epithelium, Nuclei and Oral Epithelial Dysplasia Grading
1 Introduction
2 Dataset and Annotations
3 Methodology
4 Experiment Design and Results
4.1 Full Epithelium Segmentation
4.2 Nuclear Segmentation and Feature Extraction
4.3 OED Grade Prediction
4.4 Nuclear Analysis
5 Conclusion and Future Work
References
STAMP: A Self-training Student-Teacher Augmentation-Driven Meta Pseudo-Labeling Framework for 3D Cardiac MRI Image Segmentation
1 Introduction
2 Methodology
2.1 STAMP Model Framework
2.2 Data Augmentation Strategies
2.3 Experiments
2.4 Evaluation
3 Results and Discussion
3.1 Image Segmentation Evaluation
3.2 Ablation Study
4 Conclusion
References
Implicit U-Net for Volumetric Medical Image Segmentation
1 Introduction
2 Methods
3 Experiments and Results
3.1 Experimental Set-Up
3.2 Results and Discussion
4 Conclusion
References
A Deep-Learning Lesion Segmentation Model that Addresses Class Imbalance and Expected Low Probability Tissue Abnormalities in Pre and Postoperative Liver MRI
1 Introduction
2 Methodology
2.1 Data and Pre-processing
2.2 The Network Architecture
2.3 Asymmetric Focal Loss
2.4 Training and Evaluation
2.5 Hysteresis Thresholding
2.6 Post-processing
2.7 Manual Editing
2.8 Statistical Analysis
3 Results
3.1 Hyper-parameter Tuning on the Validation Set
3.2 Validation Set Performance
3.3 Test Set Performance
4 Discussion
5 Conclusion
References
Utility of Equivariant Message Passing in Cortical Mesh Segmentation
1 Introduction
2 Background
2.1 Graph Neural Networks
2.2 E(n) Equivariance
2.3 Point Cloud Registration
2.4 Related Work
3 Methodology
3.1 Dataset and Preprocessing
3.2 Models, Training and Evaluation
3.3 Experiments
4 Conclusion
References
A Novel Framework for Coarse-Grained Semantic Segmentation of Whole-Slide Images
1 Introduction
2 The Proposed Method
2.1 Network Input
2.2 Network Architecture
2.3 Weighted Sparse Loss
2.4 Model Training
3 Experiments with Downstream Analysis Tasks
3.1 OED Grading
3.2 HNSCC Survival Analysis
4 Datasets and Experiments
4.1 Datasets
4.2 Experiments
5 Results
5.1 Pixel-wise vs Coarse Segmentation
5.2 Patch-Based Classification vs Coarse Segmentation
5.3 Inference Time Comparison
5.4 Downstream Tasks
6 Ablation Studies
6.1 Network Variations
6.2 Mini-Patch Variation
7 Conclusions
References
Generative Adversarial Network, Transformer and New Models
How Effective is Adversarial Training of CNNs in Medical Image Analysis?
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Dataset
3.2 CNN Model and Adversarial Attacks
3.3 Experiments
4 Results and Discussion
5 Conclusion
References
A U-Net Based Progressive GAN for Microscopic Image Augmentation
1 Introduction
2 Related Work
2.1 GANs for Image Synthesis
2.2 Medical Image Augmentation
3 U-Net Based Progressive GAN
3.1 Skip Connections
3.2 Noise Injection
4 Experiments
4.1 Datasets and Implementation Details
4.2 Evaluation
5 Conclusion
References
A Deep Generative Model of Neonatal Cortical Surface Development
1 Introduction
2 Methods
2.1 Model Architecture
3 Experimental Methods and Results
3.1 Data
3.2 Experiment 1: Cortical Maturity
3.3 Experiment 2: Prematurity
4 Conclusion
References
A Generative Framework for Predicting Myocardial Strain from Cine-Cardiac Magnetic Resonance Imaging
1 Introduction
2 Dataset and Pre-processing
2.1 Dataset
2.2 Data Preprocessing
3 Methodology
3.1 System Overview
3.2 Multi-channel Variational Autoencoder
3.3 Cardiac Motion Tag Tracking
4 Experiments and Results
4.1 Experimental Setup
4.2 Qualitative Evaluations
4.3 Tag Tracking
4.4 Strain Analysis
5 Conclusion
References
An Uncertainty-Aware Transformer for MRI Cardiac Semantic Segmentation via Mean Teachers
1 Introduction
2 Methodology
2.1 Semi-supervised Learning Framework
2.2 Segmentation Transformer
3 Experiments
3.1 Datasets
3.2 Training Details
3.3 Evaluation
3.4 Results
3.5 Ablation Study
4 Conclusion
A Appendix
References
SF-SegFormer: Stepped-Fusion Segmentation Transformer for Brain Tissue Image via Inter-Group Correlation and Enhanced Multi-layer Perceptron
1 Introduction
2 Related Works
2.1 Transformer-Layer Based Segmentation Method
3 The Proposed Method
3.1 Double Paired-modality Encoding (DPE) Network
3.2 Cross-Feature Decoding (CFD) Network
3.3 Semantical Double-Boundary Generation (SDBG) Branch
3.4 Segmentation Loss
4 Experimental Results and Analysis
4.1 Implementation Details
4.2 Ablation Study
4.3 Performance Comparisons
5 Conclusion
References
Polyp2Seg: Improved Polyp Segmentation with Vision Transformer
1 Introduction
2 Related Work
2.1 Hand-Crafted Approaches
2.2 Deep Learning-Based Approaches
3 Method
3.1 Multi-scale Feature Representation
3.2 Feature Aggregation Module (FAM)
3.3 Multi-context Attention Module (MCAM)
3.4 Loss Function
4 Experimental Results
4.1 Datasets
4.2 Evaluation Metrics
4.3 Implementation Details
4.4 Quantitative Analysis
4.5 Qualitative Analysis
5 Conclusions
References
Multi-resolution Fine-Tuning of Vision Transformers
1 Introduction
2 Method
2.1 Image Dataset Selection
2.2 Multi-resolution Fine-Tuning for Transfer Learning
2.3 Fold Selection and PyTorch Model Training
3 Results
3.1 ViT 3842 vs ViT 5442 Model Comparison with no Prior Training
3.2 Multi-resolution Fine Tuning
4 Discussion
5 Conclusion
References
From Astronomy to Histology: Adapting the FellWalker Algorithm to Deep Nuclear Instance Segmentation
1 Introduction
2 Materials and Methods
2.1 Dataset Description and Evaluation Metrics
2.2 FellWalker Algorithm
2.3 Deep Learning-Based Nuclear Segmentation
2.4 Implementation Details
3 Results
3.1 Algorithm Performances Across Different Dataset Sizes
3.2 FellWalker Performance with Different Segmentation Models
4 Discussion and Conclusions
References
Image Classification
Leveraging Uncertainty in Deep Learning for Pancreatic Adenocarcinoma Grading
1 Introduction
2 Proposed Method
2.1 Dataset
2.2 Approximate Bayesian Neural Networks (BCNN) and Model Uncertainty
2.3 Uncertainty Metrics in Deep Learning
2.4 Experiment
3 Model Performance
4 Prediction Error vs Uncertainty
4.1 Bayesian Model Uncertainty
4.2 Relationship Between the Accuracy and Uncertainty
4.3 Performance Improvement via Uncertainty-Aware Cancer Grading Classification
5 Leveraging Uncertainty in Classification Error and Reject Tradeoff
5.1 Evaluation Metric - Accuracy-Rejection Quotient (ARQ)
6 Conclusion and Future Work
References
A Novel Bi-level Lung Cancer Classification System on CT Scans
1 Introduction
1.1 Contribution of the Work
1.2 Related Work
2 Dataset
3 Proposed Methodology
3.1 Overview
3.2 Image Preprocessing
3.3 Data Augmentation
3.4 Level-1 Classification: CBWO-CNN
3.5 Level-2 Classification: SE-Xception
4 Results and Discussion
4.1 Level-1 Classification
4.2 Level-2 Classification
5 Conclusion
References
Jointly Boosting Saliency Prediction and Disease Classification on Chest X-ray Images with Multi-task UNet
1 Introduction
2 Background
2.1 Saliency Prediction with Deep Learning
2.2 CXR Image Classification with Deep Learning
3 Multi-task Learning Method
3.1 Multi-task UNet
3.2 Multi-task Training Scheme
4 Dataset and Evaluation Methods
5 Experiments and Result
5.1 Benchmark Comparison
5.2 Ablation Study
6 Discussion
A Mathematical Derivation of Vicious Circle for Overfitting
B Training Settings
C Saliency Map visualization
References
Deep Bayesian Active-Learning-to-Rank for Endoscopic Image Data
1 Introduction
2 Related Work
3 Deep Bayesian Active-Learning-to-Rank
3.1 Overview
3.2 Details
4 Experiments and Results
4.1 Dataset
4.2 Implementation Details
4.3 Baselines
4.4 Evaluation of Relative Severity Estimation
4.5 Relationship Between Uncertainty and Class Imbalance
4.6 Application Using the Estimated Rank Score
5 Conclusion
References
Improving Image Representations via MoCo Pre-training for Multimodal CXR Classification
1 Introduction
2 Method
2.1 Model
2.2 Self-supervised Image Pre-training
3 Experiments and Results
3.1 Experimental Setup
3.2 Comparison of Self-supervised Pre-training Strategies
3.3 Model Explainability
4 Conclusion
A Per-Class Results
References
Multi-scale Graph Neural Networks for Mammography Classification and Abnormality Detection
1 Introduction
2 Methods
2.1 Multi-scale Graph Generation
2.2 Node Features Extraction
2.3 Graph Convolutional Neural Network
2.4 Implementation Details
3 Experimental Results
3.1 Experimental Setup
3.2 Evaluation of the Proposed Framework
3.3 A Finer Lesion Segmentation with Superpixels
4 Conclusion
References
TransSLC: Skin Lesion Classification in Dermatoscopic Images Using Transformers
1 Introduction
2 Methods and Materials
2.1 Image Transformer
2.2 Model Implementation Setup
2.3 Model Evaluation
2.4 Dataset
3 Experimental Results
4 Conclusion
References
Image Enhancement, Quality Assessment, and Data Privacy
Privacy Preserving and Communication Efficient Information Enhancement for Imbalanced Medical Image Classification
1 Introduction
2 Data Description and Problem Setup
2.1 Target Dataset
2.2 Source Dataset
2.3 Problem Setup
3 Methods
3.1 Transfer Learning
3.2 Generative Models and Artificial Sample
4 Experiments and Numerical Results
4.1 Classification Model
4.2 Training from Scratch on Imbalanced Dataset
4.3 Information Enhancement by the Source Dataset
4.4 Information Enhancement by Artificial Sample
4.5 Privacy Preserving and Communication Efficient Information Enhancement Procedure
4.6 Comparison and Analysis of Experiment Results
5 Conclusion
References
Contrastive Pretraining for Echocardiography Segmentation with Limited Data
1 Introduction
2 Related Work
2.1 Segmentation Networks
2.2 Ventricular Segmentation
2.3 Contrastive Learning
3 Methods
3.1 Datasets
3.2 Experimental Setup
3.3 EchoNet-Dynamic Experiments
3.4 CAMUS Experiments
4 Results
4.1 EchoNet-Dynamic
4.2 CAMUS
5 Discussion
6 Conclusion
References
High-Quality 4D-CBCT Imaging from Single Routine Scan
1 Introduction
2 Materials and Methods
3 Discussions
4 Conclusions
References
Non-iterative Blind Deblurring of Digital Microscope Images with Spatially Varying Blur
1 Introduction
2 Related Work
3 Image Blurring Model
4 Proposed Method
4.1 Determination of the Pixels of Interest
4.2 Spatially-Varying PSF Estimation
4.3 Deconvolution Filter Design
4.4 Deconvolution and Final Image Composition
5 Experiment Results
5.1 Quantitative Analysis of Blur Map Estimation
5.2 Qualitative Analysis of Deblurring
5.3 Mask and -map Results
5.4 Influence of the Selection of
5.5 Influence of the Upper Limit for -map
6 Conclusion
References
Low-Effort Re-identification Techniques Based on Medical Imagery Threaten Patient Privacy
1 Introduction
2 Related Work on Person Identification Using Medical Signals
3 Data Selection and Preparation
3.1 Endoscopic Datasets
3.2 Brain MRI Datasets
4 Implementation and Results
4.1 Feature Extraction Network
5 Conclusion
References
Removing Specular Reflection in Multispectral Dermatological Images Using Blind Source Separation
1 Introduction
2 Proposed Method
3 Results
3.1 Real Dermatological Images
3.2 Artificial Dermatological Images
4 Conclusion
References
A Multi-scale Self-supervision Method for Improving Cell Nuclei Segmentation in Pathological Tissues
1 Introduction
2 Methods
2.1 Dataset
2.2 Self-supervision Methods
2.3 Segmentation
2.4 Evaluation Metrices
3 Experiments and Results
3.1 Experiments Setup
3.2 Self Supervision Configurations
3.3 Dataset 1 Results
3.4 Dataset 2 Results
4 Discussion
5 Conclusions
References
Radiomics, Predictive Models, and Quantitative Imaging
Correlation Between IBSI Morphological Features and Manually-Annotated Shape Attributes on Lung Lesions at CT
1 Introduction
2 Materials and Methods
2.1 Study Population
2.2 Shape Features
2.3 Statistical Analysis
2.4 Implementation, Execution and Reproducible Research
3 Results
4 Discussion
5 Limitations and Future Work
6 Conclusions
References
Large-Scale Patch-Wise Pathological Image Feature Dataset with a Hardware-agnostic Feature Extraction Tool
1 Introduction
2 Method
2.1 Tiling and Feature Extraction
2.2 Containerization
2.3 Downstream Weakly Supervised Segmentation
3 Data and Experiments
3.1 PathFeature Dataset
3.2 Weakly Supervised Segmentation
4 Results
4.1 Efficiency Analysis
4.2 Efficacy Analysis
5 Conclusion
References
Predicting Myocardial Infarction Using Retinal OCT Imaging
1 Introduction
2 Methodology
2.1 Variational Autoencoder
3 Experiments and Results
3.1 Data Characteristics
3.2 Experimental Settings
3.3 Results
4 Discussion
5 Conclusion
References
On the Feasibility of Radiomic Analysis for the Detection of Breast Lesions in Speed-of-Sound Images of the Breast
1 Introduction
2 Materials and Methods
2.1 Virtual Breast Cohort
2.2 Speed-of-Sound Images
2.3 Digital Mammography Simulation
2.4 Lesion Detection
3 Experiments and Results
3.1 Performance at the Slice Level
3.2 Performance at the Breast Level
4 Discussion
5 Conclusions
References
Oral Dental Diagnosis Using Deep Learning Techniques: A Review
1 Introduction
2 Materials and Methods
2.1 Review Questions
2.2 Search Strategy
2.3 Inclusion and Exclusion Criteria
2.4 Study Selection
2.5 Data Extraction
2.6 Accuracy Measures
3 Results
3.1 Selection of the Study
3.2 Relevant Data About the Image Data Set
3.3 Architectures of the Collected Studies
3.4 Different Imaging Types in the Collected Studies
3.5 Data Characteristics
4 Key Studies
4.1 Dental Caries Detection
4.2 Oral Cancer
4.3 Multiple Disease
4.4 Detecting the Extent of Periodontal Bone Lose
4.5 Detecting Lesions
4.6 Other Diseases
5 Discussion
6 Limitations
References
Computational Image Analysis Techniques, Programming Languages and Software Platforms Used in Cancer Research: A Scoping Review
1 Introduction
2 Materials and Methods
2.1 Research Questions
2.2 Keywords and Data Source
2.3 Mining Strategy
3 Results and Discussion
4 Conclusions
References
Image-Guided Intervention
A User Interface for Automatic Polyp Detection Based on Deep Learning with Extended Vision
1 Introduction
2 Data
2.1 Animal Data
2.2 Human Data
3 Methods
3.1 Video Processing System
3.2 Endoscope Assembly
3.3 Polyp Detection System Using AI
3.4 Animal Model
4 Results
5 Discussion
5.1 Limitations
6 Conclusion
References
Using Deep Learning on X-ray Orthogonal Coronary Angiograms for Quantitative Coronary Analysis
1 Introduction
2 Background
2.1 Image Processing Tools
2.2 U-Net
2.3 QCA Basis
3 Method
3.1 Image Acquisition
3.2 Coronary Segmentation
3.3 Applications
4 Results
4.1 Coronary Segmentation
4.2 Applications
5 Conclusion
References
Efficient Pipeline for Rapid Detection of Catheters and Tubes in Chest Radiographs
1 Introduction
2 Methods
2.1 Dataset
2.2 Data Preprocessing
2.3 Experimental Analysis
3 Results
4 Discussion
5 Conclusion
References
FCN-Transformer Feature Fusion for Polyp Segmentation
1 Introduction
2 FCBFormer
2.1 Transformer Branch (TB)
2.2 Fully Convolutional Branch (FCB)
2.3 Prediction Head (PH)
3 Experiments
3.1 Implementation Details
3.2 Evaluation
4 Conclusion
References
Correction to: Faster Diffusion Cardiac MRI with Deep Learning-Based Breath Hold Reduction
Correction to: Chapter 8 in: G. Yang et al. (Eds.): Medical Image Understanding and Analysis, LNCS 13413, https://doi.org/10.1007/978-3-031-12053-4_8
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Medical Image Understanding and Analysis

Description

More details

Other editions

Additional editions

Content

System requirements