Medical Image Computing and Computer Assisted Intervention – MICCAI 2023: 26th International Conference, Vancouver, BC, Canada, October 8–12, 2023, ... Part I (Lecture Notes in Computer Science)

✍ Scribed by Hayit Greenspan; Anant Madabhushi; Parvin Mousavi; Septimiu Salcudean; James Duncan; Tanveer Syeda-Mahmood; Russell Taylor

• Tongue: English
• Leaves: 821
• Category: Library

✦ Table of Contents

Preface
Organization
Contents – Part I
Machine Learning with Limited Supervision
PET-Diffusion: Unsupervised PET Enhancement Based on the Latent Diffusion Model
1 Introduction
2 Method
2.1 Image Compression
2.2 Latent Diffusion Model
2.3 Implementation Details
3 Experiments
3.1 Dataset
3.2 Ablation Analysis
3.3 Comparison with State-of-the-Art Methods
3.4 Generalization Evaluation
4 Conclusion and Limitations
References
MedIM: Boost Medical Image Representation via Radiology Report-Guided Masking
1 Introduction
2 Approach
2.1 Image and Text Encoders
2.2 Report-Guided Mask Generation
2.3 Decoder for Reconstruction
2.4 Objective Function
2.5 Downstream Transfer Learning
3 Experiments and Results
3.1 Experimental Details
3.2 Comparisons with Different Pre-training Methods
3.3 Discussions
4 Conclusion
References
UOD: Universal One-Shot Detection of Anatomical Landmarks
1 Introduction
2 Method
2.1 Stage I: Contrastive Learning
2.2 Stage II: Supervised Learning
3 Experiment
3.1 Experimental Results
4 Conclusion
References
S2ME: Spatial-Spectral Mutual Teaching and Ensemble Learning for Scribble-Supervised Polyp Segmentation
1 Introduction
2 Methodology
2.1 Preliminaries
2.2 S2ME: Spatial-Spectral Mutual Teaching and Ensemble Learning
3 Experiments
3.1 Experimental Setup
3.2 Results and Analysis
3.3 Ablation Studies
4 Conclusion
References
Modularity-Constrained Dynamic Representation Learning for Interpretable Brain Disorder Analysis with Functional MRI
1 Introduction
2 Materials and Methodology
2.1 Subjects and Image Preprocessing
2.2 Proposed Method
3 Experiment
4 Discussion
5 Conclusion and Future Work
References
Anatomy-Driven Pathology Detection on Chest X-rays
1 Introduction
2 Related Work
3 Method
3.1 Model
3.2 Inference
3.3 Training
3.4 Dataset
4 Experiments and Results
4.1 Experimental Setup and Baselines
4.2 Pathology Detection Results
5 Discussion and Conclusion
References
VesselVAE: Recursive Variational Autoencoders for 3D Blood Vessel Synthesis
1 Introduction
2 Methods
3 Experimental Setup
4 Results
5 Conclusions
References
Dense Transformer based Enhanced Coding Network for Unsupervised Metal Artifact Reduction
1 Introduction
2 Methodology
2.1 Hierarchical Disentangling Encoder (HDE)
2.2 Dense Transformer for Disentanglement (DTD)
2.3 Second-Order Disentanglement for MA Reduction (SOD-MAR)
2.4 Loss Function
3 Empirical Results
3.1 Ablation Study
3.2 Comparison to State-of-the-Art (SOTA)
4 Conclusion
References
Multi-scale Cross-restoration Framework for Electrocardiogram Anomaly Detection
1 Introduction
2 Method
2.1 Multi-scale Cross-restoration
2.2 Anomaly Score Measurement
3 Experiments
3.1 Comparisons with State-of-the-Arts
3.2 Ablation Study and Sensitivity Analysis
4 Conclusion
References
Correlation-Aware Mutual Learning for Semi-supervised Medical Image Segmentation
1 Introduction
2 Method
2.1 Overview
2.2 Cross-Sample Mutual Attention Module
2.3 Omni-Correlation Consistency Regularization
3 Experiments and Results
4 Conclusion
References
TPRO: Text-Prompting-Based Weakly Supervised Histopathology Tissue Segmentation
1 Introduction
2 Method
2.1 Classification with Deep Text Guidance
2.2 Knowledge Attention Module
2.3 Pseudo Label Generation
3 Experiments
3.1 Dataset
3.2 Implementation Details
3.3 Compare with State-of-the-Arts
3.4 Ablation Study
4 Conclusion
References
Additional Positive Enables Better Representation Learning for Medical Images
1 Introduction
2 Related Work
3 Method
3.1 Framework Overview
3.2 Additional Positive Selection Using TracIn
4 Experiments and Results
4.1 Experimental Setups
4.2 Semi-supervised Learning
4.3 Transfer Learning
5 Conclusion
References
Multi-modal Semi-supervised Evidential Recycle Framework for Alzheimer's Disease Classification
1 Introduction
2 Methods
2.1 Original Deep Evidential Regression (DER)
2.2 Evidential Regression Beyond DER
2.3 Model and Workflow
3 Experiments and Results
4 Conclusions
References
3D Arterial Segmentation via Single 2D Projections and Depth Supervision in Contrast-Enhanced CT Images
1 Introduction
2 Related Work
3 Methodology
4 Experimental Design
5 Results
6 Conclusion
References
Automatic Retrieval of Corresponding US Views in Longitudinal Examinations
1 Introduction
2 Method
2.1 Problem Formulation
2.2 Contrastive Learning Framework for Muscle View Matching
2.3 The Model Architecture
3 Materials
4 Experiments and Results
4.1 Implementation Details
4.2 Results
5 Discussion and Conclusion
References
Many Tasks Make Light Work: Learning to Localise Medical Anomalies from Multiple Synthetic Tasks
1 Introduction
2 Method
3 Experiments and Results
4 Conclusion
References
AME-CAM: Attentive Multiple-Exit CAM for Weakly Supervised Segmentation on MRI Brain Tumor
1 Introduction
2 Attentive Multiple-Exit CAM (AME-CAM)
3 Experiments
3.1 Dataset
3.2 Implementation Details and Evaluation Protocol
4 Results
4.1 Quantitative and Qualitative Comparison with State-of-the-Art
4.2 Ablation Study
5 Conclusion
References
Cross-Adversarial Local Distribution Regularization for Semi-supervised Medical Image Segmentation
1 Introduction
2 Method
2.1 The Minimax Optimization of VAT
2.2 Adversarial Local Distribution
2.3 Cross-Adversarial Distribution Regularization
2.4 Multiple Particle-Based Search to Approximate the Cross-ALD Regularization
2.5 Cross-ALD Regularization Loss in Medical Semi-supervised Image Segmentation
3 Experiments
3.1 Diversity of Adversarial Particle Comparison
3.2 Performance Evaluation on the ACDC and la Datasets
3.3 Ablation Study
4 Conclusion
References
AMAE: Adaptation of Pre-trained Masked Autoencoder for Dual-Distribution Anomaly Detection in Chest X-Rays
1 Introduction
2 Method
2.1 Stage 1- Proxy Task to Detect Synthetic Anomalies
2.2 Stage 2- MAE Inter-Discrepancy Adaptation
3 Experiments
4 Conclusion
References
Gall Bladder Cancer Detection from US Images with only Image Level Labels
1 Introduction
2 Datasets
3 Our Method
4 Experiments and Results
5 Conclusion
References
Dual Conditioned Diffusion Models for Out-of-Distribution Detection: Application to Fetal Ultrasound Videos
1 Introduction
2 Related Work
3 Methods
3.1 Dual Conditioned Diffusion Models
3.2 Dual Conditioning Mechanism
3.3 In-Distribution Classifier
4 Experiments and Results
4.1 Results
4.2 Ablation Study
5 Conclusion
References
Weakly-Supervised Positional Contrastive Learning: Application to Cirrhosis Classification
1 Introduction
2 Method
3 Experiments
3.1 Datasets
3.2 Architecture and Optimization
4 Results and Discussion
5 Conclusion
References
Inter-slice Consistency for Unpaired Low-Dose CT Denoising Using Boosted Contrastive Learning
1 Introduction
2 Method
2.1 Contrastive Learning for Unpaired Data
2.2 Contrastive Learning for Inter-slice Consistency
2.3 Boosted Contrastive Learning
3 Experiments
3.1 Dataset and Training Details
3.2 Comparison of Different Methods
3.3 Line Plot over Slices
3.4 Discussion
4 Conclusion
References
DAS-MIL: Distilling Across Scales for MIL Classification of Histological WSIs
1 Introduction
2 Related Work
3 Method
4 Experiments
4.1 Comparison with the State-of-the-art
4.2 Model Analysis
5 Conclusion
References
SLPD: Slide-Level Prototypical Distillation for WSIs
1 Introduction
2 Method
2.1 Overview
2.2 Preliminaries
2.3 Slide-Level Clustering
2.4 Intra-Slide Distillation
2.5 Inter-Slide Distillation
3 Experimental Results
3.1 Weakly-Supervised Classification
3.2 Ablation Study
4 Conclusion
References
PET Image Denoising with Score-Based Diffusion Probabilistic Models
1 Introduction
2 Method
2.1 Training Stage
2.2 Sampling Stage
3 Experiment
3.1 Experimental Setup
3.2 Experimental Results
4 Conclusion
References
LSOR: Longitudinally-Consistent Self-Organized Representation Learning
1 Introduction
2 Method
2.1 LSOR
2.2 SOM Similarity Grid
3 Experiments
3.1 Experimental Setting
3.2 Results
4 Conclusion
References
Self-supervised Learning for Physiologically-Based Pharmacokinetic Modeling in Dynamic PET
1 Introduction
1.1 Related Work
2 Methodology and Materials
2.1 Kinetic Modelling
2.2 Proposed Pipeline
2.3 Curve Fit
2.4 Dataset
3 Results
4 Discussion
5 Conclusion
References
Geometry-Invariant Abnormality Detection
1 Introduction
2 Background
2.1 Vector-Quantized Variational Autoencoder
2.2 Transformer
2.3 Anomaly Detection via Kernel Density Estimation Maps
3 Method
3.1 VQ-VAE Spatial Conditioning
3.2 Transformer Spatial Conditioning
3.3 Data
4 Results
5 Conclusion
References
Modeling Alzheimers' Disease Progression from Multi-task and Self-supervised Learning Perspective with Brain Networks
1 Introduction
2 Method
2.1 Problem Formalization
2.2 Overview
2.3 Self-supervised Spatio-Temporal Representation Learning
2.4 Temporal Multi-task Learning
3 Experiments
3.1 Dataset and Experimental Settings
3.2 Effectiveness Evaluation
3.3 Discussion
4 Conclusion
References
Unsupervised Discovery of 3D Hierarchical Structure with Generative Diffusion Features
1 Introduction
2 Background on Diffusion Models
3 Method
4 Experiments
4.1 Datasets
4.2 Implementation Details
4.3 Results
5 Conclusion
References
Domain Adaptation for Medical Image Segmentation Using Transformation-Invariant Self-training
1 Introduction
2 Methodology
2.1 Model
2.2 Training
3 Experimental Setup
4 Results
5 Conclusion
References
Multi-IMU with Online Self-consistency for Freehand 3D Ultrasound Reconstruction
1 Introduction
2 Methodology
2.1 Modal-Level Self-supervised Strategy
2.2 Sequence-Level Self-consistency Strategy
3 Experiments
4 Conclusion
References
Deblurring Masked Autoencoder Is Better Recipe for Ultrasound Image Recognition
1 Introduction
2 Method
2.1 Preliminary: MAE
2.2 Our Proposed Deblurring MAE
3 Experiments and Results
3.1 Experimental Settings
3.2 Results and Comparisons
4 Conclusion and Future Work
References
You've Got Two Teachers: Co-evolutionary Image and Report Distillation for Semi-supervised Anatomical Abnormality Detection in Chest X-Ray
1 Introduction
2 Method
3 Experiments
4 Conclusion
References
Masked Vision and Language Pre-training with Unimodal and Multimodal Contrastive Losses for Medical Visual Question Answering
1 Introduction
2 Methods
2.1 Model Architecture
2.2 Unimodal and Multimodal Contrastive Losses
2.3 Image Text Matching
2.4 Masked Language Modeling
2.5 Masked Image Strategy
3 Experiments
3.1 Datasets
3.2 Implementation Details
3.3 Comparison with the State-of-the-Arts
3.4 Ablation Study
3.5 Visualization
4 Conclusion
References
CL-ADDA: Contrastive Learning with Amplitude-Driven Data Augmentation for fMRI-Based Individualized Predictions
1 Introduction
2 Method
2.1 Overall Workflow of CL-ADDA
2.2 Amplitude-Driven Data Augmentation
2.3 Contrastive Learning on Functional Connectivity Maps
2.4 Individualized Prediction and Loss Function
3 Experiments and Results
3.1 Dataset
3.2 The Performance of CL-ADDA
3.3 Comparison Experiments
3.4 Ablation Experiments
4 Conclusion
References
An Auto-Encoder to Reconstruct Structure with Cryo-EM Images via Theoretically Guaranteed Isometric Latent Space, and Its Application for Automatically Computing the Conformational Pathway
1 Introduction
2 Related Work
2.1 Existing Reconstruction Methods
2.2 RaDOGAGA Revisit
2.3 Differences Between cryoTWIN and Existing Methods
3 Proposed Method
3.1 cryoTWIN
3.2 Analysis of CryoTWIN
3.3 Computation for Conformational Pathway
4 Numerical Experiment
5 Conclusion and Future Work
References
Knowledge Boosting: Rethinking Medical Contrastive Vision-Language Pre-training
1 Introduction
2 Methodology
2.1 Framework Formulation
2.2 Knowledge Semantic Enhancement
2.3 Knowledge Semantic Guidance
3 Experiment
4 Conclusion
References
A Small-Sample Method with EEG Signals Based on Abductive Learning for Motor Imagery Decoding
1 Introduction
2 Problem Definition and Method
2.1 Problem Definition
2.2 Architecture of the SSE-ABL Framework
2.3 Sample Representation
2.4 Multiscale Feature Fusion
2.5 Motion Intention Estimation
3 Results
4 Conclusion
References
Multi-modal Variational Autoencoders for Normative Modelling Across Multiple Imaging Modalities
1 Introduction
2 Methods
3 Experiments
4 Discussion and Further Work
References
LOTUS: Learning to Optimize Task-Based US Representations
1 Introduction
2 Methodology
2.1 Differentiable Ultrasound Renderer
2.2 End-to-End Learning
3 Experimental Setup
4 Results and Discussion
5 Conclusion
References
Unsupervised 3D Out-of-Distribution Detection with Latent Diffusion Models
1 Introduction
2 Methods
2.1 Latent Diffusion Models
2.2 OOD Detection with LDMs
2.3 Spatial Anomaly Maps
3 Experiments
3.1 Data
3.2 Implementation Details
4 Results and Discussion
5 Conclusion
References
Improved Multi-shot Diffusion-Weighted MRI with Zero-Shot Self-supervised Learning Reconstruction
1 Introduction
2 Method
2.1 PI Techniques for dMRI
2.2 Network Design
2.3 Zero-Shot Self-supervised Learning
2.4 Experiment Details
3 Results
4 Discussion and Conclusion
References
Infusing Physically Inspired Known Operators in Deep Models of Ultrasound Elastography
1 Introduction
2 Materials and Methods
2.1 PICTURE
2.2 Known Operators
2.3 Unsupervised Training
2.4 Dataset and Quantitative Metrics
2.5 Network Architecture and Training
3 Results and Discussions
3.1 Compared Methods
3.2 Results and Discussions
4 Conclusions
References
Weakly Supervised Lesion Localization of Nascent Geographic Atrophy in Age-Related Macular Degeneration
1 Introduction
2 Methods
2.1 Dataset
2.2 Deep Learning Architecture
2.3 Model Training, Tuning, and Validation Test
3 Results
3.1 Performance and Saliency Map Analysis of the nGA Classification Model on OCT Volumes
3.2 Performance of the Weakly Supervised Localization of nGA Lesions
4 Conclusion and Discussion
References
Can Point Cloud Networks Learn Statistical Shape Models of Anatomies?
1 Introduction
2 Background
3 Methods
3.1 Point Completion Networks for SSM
3.2 Evaluation Metrics
4 Experiments
5 Discussion and Conclusion
References
CT-Guided, Unsupervised Super-Resolution Reconstruction of Single 3D Magnetic Resonance Image
1 Introduction
2 Methodology
2.1 Super-Resolution Network (SRNet)
2.2 Cross-Modality Image Translation Network (CITNet)
2.3 Training Strategy
3 Experiments
4 Conclusion
References
Image2SSM: Reimagining Statistical Shape Models from Images with Radial Basis Functions
1 Introduction
2 Methods
2.1 Representing Shapes Using RBFs
2.2 Loss Functions
3 Results
3.1 Datasets
4 Conclusion
References
MDA-SR: Multi-level Domain Adaptation Super-Resolution for Wireless Capsule Endoscopy Images
1 Introduction
2 Methods
2.1 Overview of the Proposed Method
2.2 Adaptive Degradation
2.3 Domain Adaptation SR
3 Experiments
3.1 Experiment Settings
3.2 Training Details
3.3 Results and Discussions
4 Conclusion
References
PROnet: Point Refinement Using Shape-Guided Offset Map for Nuclei Instance Segmentation
1 Introduction
2 Methodology
2.1 Loss Functions Using Pseudo Labels
2.2 Refinement via Expectation Maximization Algorithm
3 Experiments
4 Conclusion
References
Self-Supervised Domain Adaptive Segmentation of Breast Cancer via Test-Time Fine-Tuning
1 Introduction
2 Methodology
2.1 Test-Time Fine-Tuning (TTFT) Network and Its Pipeline
2.2 Parameter Fluctuation: Parameter Randomization Method
3 Experiments
3.1 Experimental Set-Ups
3.2 Comparison Analysis
3.3 Ablation Study
4 Discussion and Conclusion
References
Decoupled Consistency for Semi-supervised Medical Image Segmentation
1 Introduction
2 Method
2.1 Dynamic Consistency Threshold
2.2 Decoupled Consistency
3 Experiment and Results
3.1 Dataset
3.2 Implementation Details
3.3 Results
4 Conclusion
References
Combating Medical Label Noise via Robust Semi-supervised Contrastive Learning
1 Introduction
2 Related Work
2.1 Conventional Methods with Noisy Labels
2.2 Semi-supervised Learning
2.3 Contrastive Learning
3 Semi-supervised Contrastive Learning
3.1 Mixup Feature Embedding
3.2 Semi-Supervised Learning
3.3 Similarity Contrastive Learning
4 Experiments
4.1 Implementation Details and Settings
4.2 Comparisons with the State-of-the-arts
4.3 Parameter Analysis and Ablation Studies and Visualizations
5 Conclusion
References
Multi-scale Self-Supervised Learning for Longitudinal Lesion Tracking with Optional Supervision
1 Introduction
2 Background and Motivation
3 Method
3.1 Problem Definition
3.2 Training Stage
3.3 Inference Stage
4 Experiments
4.1 Datasets and Setup
4.2 Evaluation
5 Conclusion
References
Tracking Adaptation to Improve SuperPoint for 3D Reconstruction in Endoscopy
1 Introduction
2 Related Work
3 Tracking Adaptation for Local Feature Learning
4 Experiments
5 Conclusions
References
Structured State Space Models for Multiple Instance Learning in Digital Pathology
1 Introduction
2 Related Work
3 Method
3.1 Neural State Space Models
3.2 MIL Training
3.3 Multitask Training
3.4 Implementation Details
4 Experiments and Discussion
4.1 Data
4.2 Results
5 Conclusions
References
vox2vec: A Framework for Self-supervised Contrastive Learning of Voxel-Level Representations in Medical Images
1 Introduction
2 Related Work
3 Method
3.1 Sampling of Positive and Negative Pairs
3.2 Architecture
3.3 Loss Function
3.4 Evaluation Protocol
4 Experiments
4.1 Pre-training
4.2 Evaluation
5 Results
6 Conclusion
References
Mesh2SSM: From Surface Meshes to Statistical Shape Models of Anatomy
1 Introduction
2 Method
2.1 Correspondence Generation
2.2 Analysis
2.3 Training
3 Experiments and Discussion
3.1 Results
3.2 Limitations and Future Scope
4 Conclusion
References
Graph Convolutional Network with Morphometric Similarity Networks for Schizophrenia Classification
1 Introduction
2 Materials and Methods
2.1 Datasets
2.2 Morphometric Similarity Networks
2.3 Graph Construction
2.4 Spectral Graph Convolutions
2.5 Interpretability
3 Experiments and Results
3.1 Results and Analysis
3.2 Ablation Studies
4 Conclusion
References
M-FLAG: Medical Vision-Language Pre-training with Frozen Language Models and Latent Space Geometry Optimization
1 Introduction
2 Methods
2.1 Frozen Language Model
2.2 Alignment and Uniformity
3 Experiments
3.1 Dataset for Pre-training
3.2 Datasets for Downstream Tasks
3.3 Results
3.4 Dimensional Collapse Analysis
3.5 Ablation Study
4 Conclusion
References
Machine Learning - Transfer Learning
Foundation Ark: Accruing and Reusing Knowledge for Superior and Robust Performance
1 Introduction
2 Accruing and Reusing Knowledge
3 Experiments and Results
3.1 Ark Outperforms SOTA Fully/Self-supervised Methods on Various Tasks for Thoracic Disease Classification
3.2 Ark Provides Generalizable Representations for Segmentation Tasks
3.3 Ark Offers Embeddings with Superior Quality over Google CXR-FM
3.4 Ark Shows a Lower False-Negative Rate and Less Gender Bias
4 Conclusions and Future Work
References
Masked Frequency Consistency for Domain-Adaptive Semantic Segmentation of Laparoscopic Images
1 Introduction
2 Method
2.1 Image Frequency Representation
2.2 Masking Strategy
2.3 Consistency Regularization
3 Experiments
3.1 Datasets and Implementation
3.2 Qualitative Evaluation
3.3 Quantitative Evaluation
4 Conclusion
References
Pick the Best Pre-trained Model: Towards Transferability Estimation for Medical Image Segmentation
1 Introduction
2 Methodology
2.1 Problem Formulation
2.2 Class Consistency with Feature Variety Constraint TE Method
3 Experiment
3.1 Experiment on MSD Dataset
3.2 Ablation Study
4 Conclusion
References
Source-Free Domain Adaptive Fundus Image Segmentation with Class-Balanced Mean Teacher
1 Introduction
2 Method
2.1 Weak-Strong Augmented Mean Teacher
2.2 Global Knowledge Guided Loss Calibration
3 Experiments
3.1 Experimental Results
3.2 Further Analyses
4 Conclusion
References
Unsupervised Domain Adaptation for Anatomical Landmark Detection
1 Introduction
2 Method
2.1 Landmark Detection Model
2.2 Landmark-Aware Self-training
2.3 Domain Adversarial Learning
3 Experiments
3.1 Experimental Settings
3.2 Results
3.3 Model Analysis
3.4 Qualitative Results
4 Conclusion
References
MetaLR: Meta-tuning of Learning Rates for Transfer Learning in Medical Imaging
1 Introduction
2 Method
2.1 Formulation of Meta Learning Rate
2.2 Online Learning Rate Adaptation
2.3 Proportional Hyper Learning Rate
2.4 Generalizability Validation on Training Data Batch
3 Experiments and Analysis
3.1 Experimental Settings
3.2 Ablation Study
3.3 Comparative Experiments
3.4 Discussion and Findings
4 Conclusion
References
Multi-Target Domain Adaptation with Prompt Learning for Medical Image Segmentation
1 Introduction
2 Methodology
2.1 Learning Domain-Specific Prompts
2.2 Learning Domain-Aware Representation by Fusion
2.3 Adversarial Learning to Enhance the Generalization Ability
2.4 Total Loss
2.5 Implementation Details
3 Experiments and Results
3.1 Datasets
3.2 Comparison with State-of-the-Art (SOTA) Method
3.3 Ablation Study
4 Conclusion
References
Spectral Adversarial MixUp for Few-Shot Unsupervised Domain Adaptation
1 Introduction
2 Methods
2.1 Domain-Distance-Modulated Spectral Sensitivity (DoDiSS)
2.2 Sensitivity-Guided Spectral Adversarial Mixup (SAMix)
3 Experiments and Results
3.1 Implementation Details
3.2 Method Effectiveness
3.3 Data Efficiency
3.4 Ablation Study
4 Discussion and Conclusion
References
Cross-Dataset Adaptation for Instrument Classification in Cataract Surgery Videos
1 Introduction
2 Related Work
3 Proposed Method
4 Experiments and Results
5 Conclusion
References
Black-box Domain Adaptative Cell Segmentation via Multi-source Distillation
1 Introduction
2 Method
3 Experiments
4 Conclusion
References
MedGen3D: A Deep Generative Framework for Paired 3D Image and Mask Generation
1 Introduction
2 Preliminary
2.1 Diffusion Probabilistic Model
2.2 Classifier-Free Guidance
3 Methodology
3.1 3D Mask Generator
3.2 Conditional Image Generator
4 Experiments and Results
4.1 Datasets and Setups
4.2 Evaluate the Quality of Synthetic Image.
4.3 Evaluate the Benefits for Segmentation Task
5 Conclusion
References
Unsupervised Domain Transfer with Conditional Invertible Neural Networks
1 Introduction
2 Materials and Methods
2.1 Domain Transfer with Conditional Invertible Neural Networks
2.2 Spectral Imaging Data
3 Experiments and Results
4 Discussion
References
Author Index