Bildverarbeitung für die Medizin 2023

Name: Bildverarbeitung für die Medizin 2023 | Proceedings, German Workshop on Medical Image Computing, Braunschweig, July 2-4, 2023
Brand: Springer Vieweg
Price: 86.99 EUR
Availability: OnlineOnly

Proceedings, German Workshop on Medical Image Computing, Braunschweig, July 2-4, 2023

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

Springer Vieweg (Verlag)

Erschienen am 1. Juni 2023

XXI, 317 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-658-41657-7 (ISBN)

86,99 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Personen

Inhalt

Intro
Bildverarbeitung für die Medizin 2023
Sponsoren und Unterstützer des Workshops BVM 2023
Preisträger der BVM 2022 in Heidelberg
Vorwort
Inhaltsverzeichnis
1 Keynote: Beyond Supervised Learning. Exploring Novel Machine Learning Approaches for Robust Medical Image Analysis
References
2 Keynote: Fully Automated Bone Removal in CBCT of the Lower Body Stem
3 Extending Tempcyclegan for Virtual Augmentation of Gastrointestinal Endoscopy Training Simulators
1 Introduction
2 Materials and methods
2.1 Proposed extensions of the tempCycleGAN
2.1.1 n-frame tempCycleGAN.
2.1.2 Conditional n-frame tempCycleGAN.
2.2 Data set and training
2.3 Experiments and evaluation
3 Results
4 Discussion
Acknowledgement.
References
4 Abstract: Fiducial Marker Recovery and Detection From Severely Truncated Data in Navigation-assisted Spine Surgery
References
5 Abstract: C-arm Positioning for Standard Projections During Spinal Implant Placement
References
6 Abstract: Shape-based Segmentation of Retinal Layers and Fluids in OCT Image Data
References
7 Extending nnU-Net Is All You Need
1 Introduction
2 Method
2.1 Optimization of nnU-Net's segmentation pipeline to AMOS2022
2.2 Inference strategy
2.3 Postprocessing
3 Results
4 Discussion
Acknowledgement.
References
8 Abstract: Liver Tumor Segmentation in Late-phase MRI using Multi-model Training and an Anisotropic U-Net
References
9 Automatic Vertebrae Segmentation in MR Volumes. A Comparison of Different Deep Learning-based Approaches
1 Introduction
2 Materials and methods
2.1 Dataset
2.2 Network 1: 3D U-net (baseline)
2.3 Network 2: iterative model
2.4 Network 3: Payer et al.
2.5 Loss functions
2.5.1 Dice loss.
2.5.2 Weighted dice loss.
2.5.3 Weighted FP and FN.
2.6 Evaluation metrics
2.6.1 Vertebrae identification.
2.6.2 Vertebrae segmentation.
3 Results
4 Discussion
References
10 Learnable Slice-to-volume Reconstruction for Motion Compensation in Fetal Magnetic Resonance Imaging
1 Introduction
2 Material and methods
2.1 Hybrid reconstruction framework
2.2 Pre-registration
2.3 Image-to-image translation
2.4 Data
2.5 Training details
3 Results
3.1 Quantitative results
3.2 Qualitative results
3.3 Expert survey
4 Discussion
Acknowledgement.
References
11 A Vesselsegmentation-based CycleGAN for Unpaired Multi-modal Retinal Image Synthesis
1 Introduction
2 Materials and methods
2.1 VesselCycleGAN for unpaired retinal image-to-image translation
2.2 Retinal datasets
2.3 Experimental details
3 Results
4 Discussion
References
12 Enhancing Medical Image Segmentation with Anatomy-aware Label Dependency
1 Introduction
2 Materials and methods
2.1 Dataset acquisition and preprocessing
2.2 Configuration 1: ventral organ segmentation with vertebrae labels
2.2.1 Dataset adjustment.
2.2.2 Network architecture.
2.3 Configuration 2: vertebrae segmentation with ventral organ labels
2.3.1 Dataset adjustment.
2.3.2 Network architecture.
3 Results
4 Discussion
Acknowledgement.
References
13 Double Grad-CAM Guidance for Improved MRI-based Pseudo-CT Synthesis
1 Introduction
1.1 Contribution
2 Materials and methods
2.1 Data
2.2 Double grad-CAM guidance
2.3 Training
2.4 Evaluation
3 Results
4 Discussion
References
14 Planning of Spherical Volumes for Treating Renal Tumors by Thermal Ablation with Tissue Shrinkage Estimation
1 Introduction
2 Materials and methods
2.1 Shrinkage modelling
2.2 Modelling the performance of two MWA systems
2.3 Radius predictive model
3 Experiments
4 Results
5 Discussion
References
15 Cerebral Vessel Tree Estimation from Non-contrast CT using Deep Learning Methods
1 Introduction
2 Materials and methods
2.1 Data
2.2 3D U-Net
2.3 Auxiliary discriminator
2.4 Experimental setup
3 Results
4 Discussion
References
16 Abstract: Trainable Joint Bilateral Filters for Enhanced Prediction Stability in Low-dose CT
References
17 Improved Tractography by Means of DL-based DWI Image Enhancement
1 Introduction
2 Materials and methods
2.1 Dataset description
2.1.1 The HCP dataset.
2.1.2 The in-house dataset.
2.2 Image enhancement with a CNN
2.2.1 Dataset preparation.
2.2.2 Network architecture.
2.2.3 Training set-up.
2.2.4 Evaluation metric.
2.3 XTRACT analysis
2.3.1 Data preprocessing.
2.3.2 Atlas generation.
2.4 Evaluation
3 Results
3.1 Image enhancement with CNN
3.2 XTRACT
4 Discussion
References
18 Deep Learning Approaches for Contrast Removal from Contrast-enhanced CT. Streamlining Personalized Internal Dosimetry
1 Introduction
2 Materials and methods
2.1 Dataset description
2.2 Preprocessing
2.3 Networks architectures
2.4 Training pipeline
2.4.1 Training: paired network.
2.4.2 Training: CycleGAN.
2.4.3 Inference.
2.5 Dosimetry
3 Results
4 Discussion
Acknowledgement.
References
19 Unsupervised Super Resolution in X-ray Microscopy using a Cycle-consistent Generative Model
1 Introduction
2 Materials and methods
2.1 Cycle-consistent generative model
2.2 Loss function
2.2.1 Adversarial loss.
2.2.2 Cycle consistency loss.
2.2.3 Perceptual loss.
2.3 Dataset and training details
3 Results
4 Discussion
References
20 McLabel. A Local Thresholding Tool for Efficient Semi-automatic Labelling of Cells in Fluorescence Microscopy
1 Introduction
2 Materials and methods
2.1 General setup
2.2 Dataset
2.3 User study
2.4 Evaluation metrics
3 Results
3.1 Annotation duration
3.2 Inter-annotator comparison
4 Discussion
References
21 WEB-ANEULYSIS. A Web-based Application for the Analysis of Aneurysm Data
1 Introduction
2 Methods and materials
2.1 Input data
2.1.1 Extraction.
2.1.2 CFD.
2.1.3 Post-processing.
2.2 Implementation
2.3 New features of web-aneulysis
3 Landmarks
3.1 Morphological descriptors
3.2 Stents
4 Results
5 Discussion
5.1 Limitations
5.2 Future work
References
22 Towards Clinical Translation of Deep Learning-based Classification of DSA Image Sequences for Stroke Treatment
1 Introduction
2 Methods
2.1 Open-source application for classification
2.2 Network architecture and model retraining
2.3 Evaluation methods
3 Results
4 Discussion
Acknowledgement.
References
23 Leveraging Semantic Information for Sonographic Wrist Fracture Assessment Within Children
1 Introduction
1.1 Dataset
2 Materials and methods
2.1 Classification network
2.2 Segmentation network
2.3 Detection network
3 Results and discussion
4 Conclusion
References
24 Applicability of BI-RADS Criteria for Deep Learning-based Classification of Suspicious Masses in Sonograms
1 Introduction
2 Materials and methods
2.1 Dataset
2.2 BI-RADS criteria and relation to image versions
2.3 Segmentation and classification models
3 Results
4 Discussion
References
25 Abstract: Tattoo-Tomographie. Freihand-3D-Photoakustik und multimodale Bildfusion
References
26 Abstract: the MIDOG Challenge 2021. Mitosis Domain Generalization in Histopathology Images
References
27 Limits of Human Expert Ensembles in Mitosis Multi-expert Ground Truth Generation
1 Introduction
2 Material and methods
2.1 Dataset D
2.2 Base classifiers C and majority vote ensembles M(C)
3 Experiments
4 Results
5 Discussion
References
28 Detection of Pulmonary Embolisms in NCCT Data Using nnDetection
1 Introduction
2 Materials and methods
2.1 Data
2.2 Methodology
2.3 Evaluation
3 Results
4 Discussion
References
29 Exploring the Effects of Contrastive Learning on Homogeneous Medical Image Data
1 Introduction
2 Materials and methods
2.1 Dataset
2.2 Partitioning and sampling strategy
2.3 Masked contrastive loss
2.4 Clustering representations
2.5 Architecture
3 Results
3.1 Validation method
3.2 Measuring performance, cohesion and uniqueness
4 Discussion
References
30 Dataset Pruning using Evolutionary Optimization
1 Introduction
2 Materials and methods
2.1 Datasets
2.2 Evolutionary optimization
2.3 Deep neural networks
3 Results
4 Discussion
Disclaimer.
References
31 Abstract: Towards Real-world Federated Learning in Medical Image Analysis using Kaapana
References
32 Compact Convolutional Transformers on Edge TPUs
1 Introduction
2 Materials and methods
2.1 Compact convolutional transformers
2.2 Edge TPU deployment
2.3 Activation-based attention map
2.4 Datasets and training
3 Results
4 Discussion
Disclaimer.
References
33 Automated Thrombus Segmentation in Stroke NCCT Incorporating Clinical Data
1 Introduction
2 Materials and methods
3 Results
4 Discussion
References
34 Geometric Deep Learning Vascular Domain Segmentation
1 Introduction
2 Materials and methods
3 Results
4 Discussion
References
35 CNN-based Whole Breast Segmentation in Longitudinal High-risk MRI Study. Initial Findings on Quality vs. Quantity of Training Annotations
1 Introduction
2 Materials and methods
3 Results
4 Discussion
5 Conclusion and future work
Acknowledgement.
References
36 Automatic Lung Nodule Segmentation in CT Imaging using an Improved 3D-Res2Unet
1 Introduction
2 Material and methods
2.1 Image data
2.2 Preprocessing
2.3 Methods
3 Results
4 Discussion
References
37 Automated Deep-learning-based Vertebral Body Localization and Instance Segmentation for Osteoporosis Assessment using CT
1 Introduction
2 Materials and methods
2.1 Algorithm
2.2 Data
2.3 Evaluation metrics
3 Results
4 Discussion
Acknowledgement.
References
38 Mitigating Unknown Bias in Deep Learning-based Assessment of CT Images DeepTechnome
1 Introduction
2 Materials and methods
2.1 DecorreLayer
2.1.1 Correlation unit.
2.1.2 Filter unit.
2.1.3 Backward pass and inference.
2.2 Data preparation and testing setup
3 Results
4 Discussion
References
39 Taming Detection Transformers for Medical Object Detection
1 Introduction
2 Materials and methods
2.1 Network architecture
2.2 Data sets
2.3 Experimental setup
3 Results
4 Discussion
Acknowledgement.
References
40 Deep Learning-based Subtyping of Atypical and Normal Mitoses using a Hierarchical Anchor-free Object Detector
1 Introduction
2 Dataset
3 Methods
3.1 Deep Learning Architecture
3.2 Training
3.3 Evaluation
4 Results
5 Discussion
References
41 Abstract: Automated Detection and Quantification of Brain Metastases on Clinical MRI Data using CNNs
References
42 Abstract: Determination of Unknown Biomechanical Parameters of a Screw-vertebra MBS Model
References
43 Challenge Results are not Reproducible
1 Introduction
2 Materials and methods
3 Results
4 Discussion
Acknowledgement.
References
44 Abstract: Is Medical Chest X-ray Data Anonymous? Deep Learning-based Patient Re-identification is Able to Exploit the Biometric Nature of Medical Chest X-ray Data
References
45 Abstract: Pan-tumor CAnine CuTaneous Cancer Histology (CATCH) Dataset
References
46 Multi-scanner Canine Cutaneous Squamous Cell Carcinoma Histopathology Dataset
1 Introduction
2 Materials and methods
2.1 Dataset validation
2.2 Technical validation
3 Results
4 Discussion
Acknowledgement.
References
47 Weakly-supervised Temporal Segmentation of Cell-cycle Stages with Center-cell Focus using Recurrent Neural Networks
1 Introduction
2 Materials and methods
2.1 Training details
2.2 Focus network loss
2.3 Classification network loss
2.4 Datasets and experiments
2.4.1 Experimental setup.
2.4.2 Evaluation criteria.
3 Results
4 Discussion
References
48 Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays
1 Introduction
2 Materials and methods
2.1 Data
2.2 Image processing
2.3 Attention-based multiple instance learning
2.4 Loss function
2.5 Baselines
2.6 Evaluation
3 Results
4 Discussion
Acknowledgement.
References
49 Deep Learning-based Automatic Assessment of AgNOR-scores in Histopathology Images
1 Introduction
2 Materials and methods
2.1 Data
2.2 Human rater experiment
2.3 AgNOR-scoring pipeline
3 Results
4 Discussion
References
50 Unsupervised detection of Small Hyperreflective Features in Ultrahigh Resolution Optical Coherence Tomography
1 Introduction
2 Materials and methods
2.1 Data acquisition and preprocessing
2.2 Algorithm description
3 Results
4 Discussion
Acknowledgement.
References
51 Comparison of CNN Architectures for Detecting Alzheimer's Disease using Relevance Maps
1 Introduction
2 Materials and methods
3 Results
4 Discussion
References
52 Abstract: Deep Geometric Supervision Improves Spatial Generalization in Orthopedic Surgery Planning
References
53 Abstract: Deep-learning on Lossily Compressed Pathology Images. Adverse Effects for ImageNet Pre-trained Models
References
54 Contrastive Representations for Unsupervised Anomaly Detection and Localization
1 Introduction
2 Method
2.1 Contrastive training
2.2 Generative model
3 Experiments
3.1 Data
3.2 Model
3.3 Training
4 Results and discussion
4.1 Discriminative power of representations for anomaly localization
4.2 Comparison to state-of-the-art anomaly localization
5 Conclusion
Acknowledgement.
References
55 Abstract: AUCMEDI. Von der Insellösung zur einheitlichen und automatischen Klassifizierung von Medizinischen Bildern
References
56 Cavity Segmentation in X-ray Microscopy Scans of Mouse Tibiae
1 Introduction
2 Materials and methods
2.1 Dataset
2.2 Thresholding-based method
2.3 Lacuane and blood vessel classification
2.4 Deep learning-based method
3 Results and discussion
4 Conclusion
Acknowledgement.
References
57 Abstract: A Spatiotemporal Model for Precise and Efficient Fully-automatic 3D Motion Correction in OCT
References
58 Label Efficient Classification in Liquid Biopsy Data by Self-supervision
1 Introduction
2 Materials and methods
2.1 Data set
2.2 Methods
2.2.1 Supervised learning:XCiTand EfficientNetB0.
2.2.2 Self-supervision: DINO.
2.3 Experiments
2.3.1 General performance in comparison.
2.3.2 Label data efficiency.
3 Results
4 Discussion
Acknowledgement.
References
59 Motif Analysis of Resting-state and Stimulus-driven fMRI Networks with Special Focus on Neurotransmitter-specific Subnetworks 2023
1 Introduction
2 Materials and methods
3 Results
4 Discussion
References
60 Detection of Arterial Occlusion on Magnetic Resonance Angiography of the Thigh using Deep Learning
1 Introduction
2 Materials and methods
2.1 Image acquisition
2.2 Labels
2.3 Neural network and training
3 Results
4 Discussion
References
61 Localizable Instruments for Navigated Treatment of Ischemic Stroke
1 Introduction
2 Methods
2.1 Trackable catheter and guidewire prototypes
2.2 Assessment of accuracy and handling
2.3 Thrombectomy navigation experiment
3 Results
4 Discussion
Acknowledgement.
References
62 Needle Tip Tracking During CT-guided Interventions using Fuzzy Segmentation
1 Introduction
2 Methodology
2.1 Data, annotation, and training setup
3 Results
4 Discussion
Disclaimer.
Acknowledgement.
References
63 Deep Learning-based Marker-less Pose Estimation of Interventional Tools using Surrogate Keypoints
1 Introduction
1.1 Related work
2 Materials and Methods
2.1 Data labeling pipeline
2.2 Needle pose estimation
2.2.1 Deep learning segmentation.
2.2.2 Triangulation.
2.3 Experiments
3 Results
4 Discussion
4.1 Conclusion
Acknowledgement.
References
64 Abstract: Implementation of a Real-time 3D-thermometry Pipeline in Gadgetron for Easy Clinical Integration
References
65 Matching Endoscopic 3D Image Data with 4D Echocardiographic Data for Extended Reality Support in Mitral Valve Repair Surgery
1 Introduction
2 Materials and methods
2.1 3D reconstruction
2.2 Registration
2.3 Implementation
2.4 Data acquisition and preprocessing
2.5 Evaluation
3 Results
4 Discussion
Acknowledgement.
References
66 Cross-modality Training Approach for CT Super-resolution Network
1 Introduction
2 Materials and methods
2.1 Datasets and pre-processing
2.1.1 Frequency-based simulation of low-resolution data.
2.1.2 Dataset.
2.2 Neural network and training
2.3 Evaluation
3 Results
3.1 Qualitative evaluation
3.1.1 Test on unseen CBCT data.
3.1.2 Test on MDCT data.
3.2 Quantitative analysis
4 Discussion
5 Conclusion
Disclaimer.
References
67 Abstract: MOOD 2020. A Public Benchmark for Out-of-distribution Detection and Localization on Medical Images
References
Correction to: McLabel. A Local Thresholding Tool for Efficient Semi-automatic Labelling of Cells in Fluorescence Microscopy
Autorenverzeichnis

Systemvoraussetzungen

Als PDF speichern Als Link merken