Bildverarbeitung für die Medizin 2020

Name: Bildverarbeitung für die Medizin 2020 | Algorithmen - Systeme - Anwendungen. Proceedings des Workshops vom 15. bis 17. März 2020 in Berlin
Brand: Springer Vieweg
Price: 49.99 EUR
Availability: OnlineOnly

Algorithmen - Systeme - Anwendungen. Proceedings des Workshops vom 15. bis 17. März 2020 in Berlin

Thomas Tolxdorff Thomas M. Deserno Heinz Handels Andreas Maier Klaus H. Maier-Hein Christoph Palm(Herausgeber*in)

Springer Vieweg (Verlag)

1. Auflage

Erschienen am 12. Februar 2020

XXIII, 351 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-658-29267-6 (ISBN)

49,99 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Bildverarbeitung für die Medizin 2020
Sponsoren des Workshops BVM 2020
Preisträger der BVM 2019 in Lübeck
Vorwort
Inhaltsverzeichnis
1 Inter-Species, Inter-Tissue Domain Adaptation for Mitotic Figure Assessment. Learning New Tricks from Old Dogs
1 Introduction
2 Material and methods
2.1 Training
3 Results
4 Discussion
Acknowledgement.
References
2 Deep Segmentation of Bacteria at Different Stages of the Life Cycle
1 Introduction
2 Methods
2.1 Network architectures
2.2 Shape-based weighting of the loss function
2.3 Network training
3 Experimental results
4 Conclusion
Acknowledgement.
References
3 Retrospective Color Shading Correction for Endoscopic Images
1 Introduction
2 Materials and methods
2.1 Gray-level shading correction using SE estimation
2.2 Color shading correction using SE estimation
2.3 Experiments
3 Results
3.1 Artificial images
3.2 Real endoscopic images
4 Discussion
References
4 Neural Network for Analyzing Prostate Cancer Tissue Microarrays. Problems and Possibilities
1 Introduction
2 Materials and methods
2.1 Data set & pre-processing
2.2 Training process and validation
3 Results
4 Discussion
References
5 Is Crowd-Algorithm Collaboration an Advanced Alternative to Crowd-Sourcing on Cytology Slides?
1 Introduction
2 Material and methods
2.1 Patch selection
2.2 Label generation
2.3 Labelling platform
2.4 Label experiment design
3 Results
4 Discussion and outlook
Acknowledgement.
References
6 Abstract: Defence of Mathematical Models for Deep Learning based Registration
References
7 Degenerating U-Net on Retinal Vessel Segmentation. What Do We Really Need?
1 Introduction
2 Materials and methods
2.1 Default U-Net configuration
2.2 Additive variants
2.3 Subtractive variants
2.4 Database description
2.5 Experimental details
3 Results
4 Discussion
References
8 COPD Classification in CT Images Using a 3D Convolutional Neural Network
1 Introduction
2 Methods
2.1 Data
2.1.1 Data Pre-processing:
3 Experiments and results
4 Discussion
5 Conclusion
Disclaimer.
References
9 Automatische Detektion von Zwischenorgan-3D-Barrieren in abdominalen CT-Daten
1 Einleitung
2 Material und Methoden
2.1 Datenvorbereitung
2.2 Gemeinsam genutzte Methoden in Training und Anwendung
2.2.1 Bounding-Box-Überschneidungen
2.3 Training der Modelle
2.3.1 Trainieren von Random-Forests für die BB-Detektion
2.3.2 Training eines U-Net für die semantische VOI-Segmentierung
2.4 Anwendung der Modelle
2.4.1 Bounding-Box und Überschneidungsdetektion
2.4.2 Partielle Segmentierung von Organen in BB-Schnittmengen
2.4.3 Barrierengenerierung
2.5 Evaluation
3 Ergebnisse
4 Diskussion
Danksagung.
Literaturverzeichnis
10 Abstract: Automatic Detection of Cervical Spine Ligaments Origin and Insertion Points
References
11 Abstract: Recognition of AML Blast Cells in a Curated Single-Cell Dataset of Leukocyte Morphologies Using Deep Convolutional Neural Networks
References
12 Fully Automated Segmentation of the Psoas Major Muscle in Clinical CT Scans
1 Introduction and previous work
2 Materials and methods
2.1 Materials
2.2 Single-pass methods
2.3 Refining the initial segmentation
2.3.1 Detecting the area of interest
2.3.2 Local refinement methods with semantic segmentation networks
2.3.3 Local refinement methods with shape contstraints
3 Experiments and results
3.1 Segmentation performance
4 Discussion
5 Conclusion
References
13 Automated Segmentation of the Locus Coeruleus from Neuromelanin-Sensitive 3T MRI Using Deep Convolutional Neural Networks
1 Introduction
2 Materials and methods
2.1 Data and manual segmentations
2.2 Network model
3 Results
4 Discussion
Acknowledgement.
References
14 Multi-Channel Volumetric Neural Network for Knee Cartilage Segmentation in Cone-Beam CT
1 Introduction
2 Materials and methods
2.1 Data
2.2 Multi-channel volumetric neural network
2.3 Loss function
2.4 Connected component analysis
3 Results
4 Discussion
Acknowledgement.
References
15 Abstract: WeLineation. STAPLE-Based Crowdsourcing for Image Segmentation
References
16 Abstract: Fully Automated Deep Learning Pipeline for Adipose Tissue Segmentation on Abdominal Dixon MRI
References
17 Semantic Lung Segmentation Using Convolutional Neural Networks
1 Introduction
1.1 Background
2 Material and method
2.1 Image database and pre-processing
2.2 Encoder-Decoder nets with atrous convolutions
2.2.1 Atrous convolutions
2.3 Training and evaluation process
3 Results
4 Discussion and conclusion
Acknowledgement.
References
18 Abstract: MITK-ModelFit. Generic Open-Source Framework for Model Fitting
References
19 Compressed Sensing for Optical Coherence Tomography Angiography Volume Generation
1 Introduction
2 Materials and methods
2.1 Algorithm
2.2 Data
3 Results
4 Discussion
Acknowledgement.
References
20 Reproduzierbare Kalibrierung von elektromagnetischen Feldverzerrungen. Experimente mit einem miniaturisierten Feldgenerator, befestigt an einer Ultraschallsonde
1 Einleitung
2 Material und Methoden
2.1 Messung der Feldverzerrung
2.2 Kalibrierungsmethode
2.3 Experimentelle Validierung
3 Ergebnisse
4 Diskussion
Danksagung.
Literaturverzeichnis
21 Deep Learning-Based Denoising of Mammographic Images Using Physics-Driven Data Augmentation
1 Introduction
2 Material and methods
2.1 Data and augmentation
2.2 Network architecture
2.2.1 Loss Function
2.2.2 Training
3 Experiments and results
3.1 Denoising of simulated noise and performance comparison
3.2 Denoising real-world noise
4 Conclusion
Disclaimer.
References
22 Video Anomaly Detection in Post-Procedural Use of Laparoscopic Videos
1 Introduction
2 Materials and methods
2.1 Adversial training in a GAN framework
2.2 Negative training phase to limit the generative abilities
3 Results
3.1 Validation
4 Discussion
References
23 Entropy-Based SVM Classifier for Automatic Detection of Motion Artifacts in Clinical MRI
1 Introduction
2 Materials and methods
2.1 Motion simulation framework and image data preparation
2.2 SVM classifier
3 Results
4 Discussion and conclusion
References
24 Tenfold your Photons. A Physically-Sound Approach to Filtering-Based Variance Reduction of Monte-Carlo-Simulated Dose Distributions
1 Introduction
2 Methods
2.1 Basic principle
2.2 Proof of concept
3 Results
4 Discussion
Disclaimer.
References
25 CT-Based Non-Destructive Quantification of 3D-Printed Hydrogel Implants
1 Introduction
2 Materials and methods
2.1 Bioprinting
2.2 Image data
2.3 Image analysis
2.3.1 Angle measurements
3 Results
4 Discussion
Acknowledgement.
References
26 Fully-Automatic CT Data Preparation for Interventional X-Ray Skin Dose Simulation
1 Introduction
2 Methods
2.1 Tissue mapping
2.2 Mask generation
2.3 Skin detection
2.4 Composition
3 Results
4 Discussion
Disclaimer.
References
27 Prediction of MRI Hardware Failures based on Image Features Using Time Series Classification
1 Introduction
2 Materials and methods
2.1 Data
2.2 Preprocessing
2.3 Classification
2.3.1 Fully convolutional neural networks
2.3.2 Residual network
2.3.3 Time convolutional neural networks
2.3.4 Long short-term memory
3 Results
4 Discussion
References
28 Prediction of MRI Hardware Failures Based on Image Features Using Ensemble Learning
1 Introduction
2 Materials and methods
2.1 Data
2.2 Preprocessing
2.3 Base learner training
2.4 Ensemble learner training
3 Results
3.1 Base learner results 1d
3.2 Base learner results matrix
3.3 Ensemble learner results
4 Discussion
References
29 Abstract: Estimation of the Principal Ischaemic Stroke Growth Directions for Predicting Tissue Outcomes
References
30 Assistive Diagnosis in Opthalmology Using Deep Learning-Based Image Retrieval
1 Introduction
2 Materials and methods
2.1 Datasets
2.2 Preprocessing
2.3 Network architecture and training
2.4 Loss functions
2.4.1 Contrastive loss
2.4.2 Triplet loss
2.4.3 Histogram loss
3 Results
4 Discussion
References
31 Multitask-Learning for the Extraction of Avascular Necrosis of the Femoral Head in MRI
1 Introduction
2 Materials and methods
2.1 Deep 3-branch multitask fully convolutional architecture
2.2 Training
2.3 Evaluation
3 Results
4 Discussion
4.1 Conclusion
References
32 Investigation of Feature-Based Nonrigid Image Registration Using Gaussian Process
1 Introduction
2 Materials and method
2.1 Feature extraction and feature matching
2.2 Estimation of dense deformation field and uncertainty map
2.2.1 GP kernel estimation
3 Results
4 Discussion
References
33 Intensity-Based 2D-3D Registration Using Normalized Gradient Fields
1 Introduction
2 Materials and methods
2.1 Digitally reconstructed radiographs
2.2 2D-3D image registration
2.3 Clinical database
2.4 Evaluation parameters
3 Results
4 Discussion
Acknowledgement.
References
34 Deep Autofocus with Cone-Beam CT Consistency Constraint
1 Introduction
2 Motion estimation and compensation framework
2.1 Autofocus
2.2 Reprojection error regression
2.3 Cone-beam consistency constraint
3 Experiments and results
4 Conclusion and discussion
Disclaimer:
References
35 Abstract: mlVIRNET. Improved Deep Learning Registration Using a Coarse to Fine Approach to Capture all Levels of Motion
References
36 Font Augmentation. Implant and Surgical Tool Simulation for X-Ray Image Processing
1 Introduction
2 Materials and methods
2.1 Font augmentation
2.2 Experiments
2.3 Dataset and training protocol
3 Results
4 Discussion
Acknowledgement.
Disclaimer.
References
37 Abstract: Segmentation of Retinal Low-Cost Optical Coherence Tomography Images Using Deep Learning
References
38 Abstract: RinQ Fingerprinting. Recurrence-Informed Quantile Networks for Magnetic Resonance Fingerprinting
References
39 Abstract: Learning to Avoid Poor Images. Towards Task-Aware C-Arm Cone-Beam CT Trajectories
References
40 Field of View Extension in Computed Tomography Using Deep Learning Prior
1 Introduction
2 Materials and methods
2.1 Deep learning artifact reduction
2.2 Data extrapolation using deep learning prior
2.3 Iterative reconstruction with TV regularization
2.4 Experimental setup
3 Results
4 Discussion
References
41 Abstract: Self-Supervised 3D Context Feature Learning on Unlabeled Volume Data
Acknowledgement.
References
42 Deep Learning Algorithms for Coronary Artery Plaque Characterisation from CCTA Scans
1 Introduction
2 Material and methods
2.1 Data
2.2 Methods
2.2.1 3D-RCNN
2.2.2 Texture-based multi-view 2D-CNN
2.2.3 2.5D-CNN
2.2.4 Modifications
2.2.5 Evaluation
3 Results
4 Discussion
Disclaimer
References
43 Abstract: Unsupervised Anomaly Localization Using Variational Auto-Encoders
References
44 Abstract: Coronary Artery Plaque Characterization from CCTA Scans Using DL and Radiomics
Disclaimer
References
45 Quantitative Comparison of Generative Shape Models for Medical Images
1 Introduction
2 Materials and methods
2.1 Statistical shape models
2.2 Locality-based multi-resolution statistical shape modeling
2.3 Autoencoders
2.4 Variational autoencoders
3 Experiments and results
3.1 Specificity and generalization
3.2 Latent space
4 Discussion and conclusion
References
46 Abstract: FastSurfer. A Fast and Accurate Deep Learning Based Neuroimaging Pipeline
References
47 VICTORIA. An Interactive Online Tool for the VIrtual Neck Curve and True Ostium Reconstruction of Intracranial Aneurysms
1 Introduction
2 Materials and methods
2.1 VICTORIA web application
2.2 Neck curve definition
2.3 Ostium creation and manipulation
3 Results
4 Discussion
Acknowledgement.
References
48 Abstract: Deep Probabilistic Modeling of Glioma Growth
References
49 Parameter Space CNN for Cortical Surface Segmentation
1 Introduction
2 Methodology
2.1 Network architecture
2.1.1 Longitude/colatitude spherical parameterization
2.1.2 Parameter space CNN (pCNN)
2.1.3 View-aggregation (p3CNN)
2.1.4 Spherical CNN
2.1.5 Mapping
2.2 Evaluation
2.2.1 Surface-based dice similarity coefficient
3 Results
4 Discussion
Acknowledgement.
References
50 Learning-Based Correspondence Estimation for 2-D/3-D Registration
1 Introduction
2 Materials and methods
2.1 Point-to-plane correspondence model-based registration
2.2 Learning-based correspondence estimation
2.2.1 General setup
2.2.2 Training procedure
2.2.3 Integration into the registration framework
2.3 Evaluation setup
2.3.1 Methodology
2.3.2 Data
3 Results
4 Discussion
References
51 Abstract: Deep Learning Based CT-CBCT Image Registration for Adaptive Radio Therapy
References
52 Learning-Based Misalignment Detection for 2-D/3-D Overlays
1 Introduction
2 Materials and methods
2.1 Registration using point-to-plane correspondences & learning correspondence weighting
2.2 Automatic misalignment detection
2.3 Data
3 Results
4 Discussion
References
53 Deep Groupwise Registration of MRI Using Deforming Autoencoders
1 Introduction
2 Materials and methods
2.1 Architecture
2.2 Training
3 Results
4 Discussion
References
54 Robust Open Field Rodent Tracking Using a Fully Convolutional Network and a Softargmax Distance Loss
1 Introduction and previous work
2 Materials and methods
2.1 Established methods for ground truth bootstrapping
2.2 Direct position regression using a VGG16 architecture
2.3 Segmentation using a u-net architecture
2.4 Direct landmark regression with a u-net and a softargmax loss
3 Experiments and results
4 Discussion and outlook
Acknowledgement.
References
55 Deep OCT Angiography Image Generation for Motion Artifact Suppression
1 Introduction
2 Materials and methods
2.1 Defect scan detection
2.2 Data set and processing
2.3 OCT-OCTA translation
3 Results
4 Discussion and conclusion
Acknowledgement.
References
56 Open Source Simulation of Fixational Eye Drift Motion in OCT Scans. Towards Better Comparability and Accuracy in Retrospective OCT Motion Correction
1 Introduction
2 Materials and methods
2.1 Random walk in a continuous domain
2.2 B-spline path representation for arbitrary sampling in time
3 Results
4 Discussion
Acknowledgement.
References
57 Abstract: Multi-Task Framework for X-Ray Guided Planning in Knee Surgery
References
58 Abstract: 3D Catheter Guidance Including Shape Sensing for Endovascular Navigation
References
59 Erlernbarkeitsstudie eines vibrotaktilen Armbands für assistive Navigation
1 Einleitung
2 Material und Methoden
3 Ergebnisse
4 Diskussion
5 Zusammenfassung und Ausblick
Danksagung.
Literaturverzeichnis
60 Visualizing the Placental Energy State in Vivo
1 Introduction
2 Materials and methods
2.1 Input data
2.2 Segmentation
2.3 Visualization
3 Results
4 Discussion
References
61 Modularization of Deep Networks Allows Cross-Modality Reuse. Lesson Learnt
1 Introduction
2 Materials and methods
2.1 Preprocessing network
2.2 Reader study
2.3 Database description
2.3.1 Fundus training database
2.3.2 OCT-A testing database
3 Results
4 Discussion
References
62 U-Net in Constraint Few-Shot Settings. Enforcing Few-Sample-Fitting for Faster Convergence of U-Net for Femur Segmentation in X-Ray
1 Introduction
2 Materials and methods
2.1 U-Net
2.2 Training
2.3 Experiments
3 Results
4 Discussion
5 Conclusion
References
63 Abstract: Multi-Scale GANs for Memory-Efficient Generation of High Resolution Medical Images
References
64 Epoch-Wise Label Attacks for Robustness Against Label Noise. Chest X-Ray Tuberculosis Classification with Corrupted Labels
1 Introduction
2 Materials and methods
2.1 Datasets
2.2 Network and training setup
2.3 Label noise
2.4 Individual label attacks
3 Results
4 Discussion
5 Conclusion
References
65 Abstract: How Big is Big Enough? A Large-Scale Histological Dataset of Mitotic Figures
Acknowledgement.
References
66 Der Einfluss von Segmentierung auf die Genauigkeit eines CNN-Klassifikators zur Mimik-Steuerung
1 Einleitung
1.1 Verwandte Arbeiten
1.2 Eigener Beitrag
2 Material und Methoden
2.1 Experimente
2.1.1 Vergleich der Klassifikatoren
2.1.2 Einfluss der Trainingsdatenmenge
3 Ergebnisse und Diskussion
3.1 Vergleich der Klassifikatoren
3.2 Einfluss der Trainingsdatenmenge
4 Zusammenfassung und Ausblick
Literaturverzeichnis
67 Imitation Learning Network for Fundus Image Registration Using a Divide-And-Conquer Approach
1 Introduction
2 Materials and methods
2.1 Fundus image registration dataset (FIRE)
2.1.1 Category A
2.1.2 Category P
2.1.3 Category S
2.2 Imitation learning network for fundus image registration
2.3 Experimental analysis of imitation learning network
2.3.1 Experiment I
2.3.2 Experiment II
3 Results
4 Discussion
References
68 Comparison of CNN Visualization Methods to Aid Model Interpretability for Detecting Alzheimer's Disease
1 Introduction
2 Materials and methods
3 Results
4 Discussion
Acknowledgement.
References
69 Abstract: Divide-And-Conquer Approach Towards Understanding Deep Networks
References
70 Abstract: Fiber Optical Shape Sensing of Flexible Instruments
References
71 Scalable HEVC for Histological Whole-Slide Image Compression
1 Introduction
1.1 Related Work
2 Materials and methods
3 Results
4 Discussion
5 Conclusion
Acknowledgement.
References
72 Image Quilting for Histological Image Synthesis
1 Introduction
1.1 Contribution
2 Materials and methods
2.1 Image quilting for histological textures
2.2 Minimum cost boundary (MCB)
2.3 Evualuation
3 Results
4 Discussion
Acknowledgement.
References
73 An Open-Source Tool for Automated Planning of Overlapping Ablation Zones. For Percutaneous Renal Tumor Treatment
1 Introduction
2 Materials and methods
2.1 Definition of uncertainty parameters
2.2 Approach for planning overlapping ablation zones
2.3 Open-source implementation
2.4 Experiments
3 Results
4 Discussion
References
74 Combining 2-D and 3-D Weight-Bearing X-Ray Images. Application to Preoperative Implant Planning in the Knee
1 Introduction
2 Methods and materials
2.1 2-D acquisition
2.2 3-D acquisition
2.3 Fusion
2.4 Evaluation
3 Results
4 Discussion
Disclaimer
References
75 Abstract: Generative Adversarial Networks for Stereoscopic Hyperrealism in Surgical Training
References
76 Haptic Rendering of Soft-Tissue for Training Surgical Procedures at the Larynx
1 Introduction
2 Materials and methods
2.1 Image data and segmentation
2.2 Scalpel model
2.3 Force-Feedback haptics
2.4 Position based dynamics
2.5 Soft tissue simulation using Nvidia flex
2.6 Haptic rendering of tool-tissue interaction
3 Results
4 Discussion
Acknowledgement.
References
Autorenverzeichnis

Systemvoraussetzungen

Als PDF speichern Als Link merken