Medical Image Computing and Computer Assisted Intervention – MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part VIII

✍ Scribed by Linwei Wang; Qi Dou; P. Thomas Fletcher; Stefanie Speidel; Shuo Li

• Publisher: Springer Nature
• Year: 2022
• Tongue: English
• Leaves: 774
• Category: Library

✦ Synopsis

The eight-volume set LNCS 13431, 13432, 13433, 13434, 13435, 13436, 13437, and 13438 constitutes the refereed proceedings of the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022, which was held in Singapore in September 2022. The 574 revised full papers presented were carefully reviewed and selected from 1831 submissions in a double-blind review process. The papers are organized in the following topical sections: Part I: Brain development and atlases; DWI and tractography; functional brain networks; neuroimaging; heart and lung imaging; dermatology; Part II: Computational (integrative) pathology; computational anatomy and physiology; ophthalmology; fetal imaging; Part III: Breast imaging; colonoscopy; computer aided diagnosis; Part IV: Microscopic image analysis; positron emission tomography; ultrasound imaging; video data analysis; image segmentation I; Part V: Image segmentation II; integration of imaging with non-imaging biomarkers; Part VI: Image registration; image reconstruction; Part VII: Image-Guided interventions and surgery; outcome and disease prediction; surgical data science; surgical planning and simulation; machine learning – domain adaptation and generalization; Part VIII: Machine learning – weakly-supervised learning; machine learning – model interpretation; machine learning – uncertainty; machine learning theory and methodologies.

✦ Table of Contents

Preface
Organization
Contents – Part VIII
Machine Learning – Weakly-Supervised Learning
CS2: A Controllable and Simultaneous Synthesizer of Images and Annotations with Minimal Human Intervention
1 Introduction
2 Methods
2.1 Unsupervised Mask Generation
2.2 Multiple AdaIN GAN
2.3 Ensemble MLP Classifier for Semantic Segmentation
3 Experiments
3.1 Dataset
3.2 Experimental Settings
3.3 Results
4 Conclusion
References
Stabilize, Decompose, and Denoise: Self-supervised Fluoroscopy Denoising
1 Introduction
2 Method
2.1 Stage 1 - Stabilize
2.2 Stage 2 - Decompose
2.3 Stage 3 - Denoise
3 Experiment
3.1 Setup Details
3.2 Results and Discussion
4 Conclusion
References
.26em plus .1em minus .1emDiscrepancy-Based Active Learning for Weakly Supervised Bleeding Segmentation in Wireless Capsule Endoscopy Images
1 Introduction
2 Methodology
2.1 CAM Generation
2.2 Discrepancy Decoder Model
2.3 CAMPUS Criterion
3 Experiments and Results
3.1 Experimental Settings
3.2 Results
4 Conclusions
References
Diffusion Models for Medical Anomaly Detection
1 Introduction
2 Method
3 Experiments
4 Results and Discussion
5 Conclusion
References
Few-Shot Generation of Personalized Neural Surrogates for Cardiac Simulation via Bayesian Meta-learning
1 Introduction
2 Methodology
2.1 Set-Conditioned Generative Model
2.2 Meta-model for Amortized Variational Inference
3 Experiments
3.1 Synthetic Experiments
3.2 Clinical Data
4 Conclusion
References
Aggregative Self-supervised Feature Learning from Limited Medical Images
1 Introduction
2 Formulation of SSL on Limited Samples
3 Multi-task Aggregative SSL
4 Self-aggregative SSL
5 Experiments
5.1 Evaluation of MT-ASSL
5.2 Evaluation of Self-ASSL
6 Conclusion
References
Learning Underrepresented Classes from Decentralized Partially Labeled Medical Images
1 Introduction
2 Problem Setup
3 Method
3.1 Preliminaries
3.2 Federated Self-supervised Learning
3.3 Energy-Based Federated Learning with Partial Labels
3.4 Prototype-Based Inference
4 Experiments
4.1 Experimental Setup
4.2 Results
5 Conclusion
References
Adversarially Robust Prototypical Few-Shot Segmentation with Neural-ODEs
1 Introduction
2 Related Works
3 Proposed Method
3.1 Problem Setting
3.2 Adversarial Training
3.3 Prototypical Neural ODE (PNODE)
4 Implementation Details
5 Experiments and Results
6 Conclusion
References
Uni4Eye: Unified 2D and 3D Self-supervised Pre-training via Masked Image Modeling Transformer for Ophthalmic Image Classification
1 Introduction
2 Methodology
2.1 Unified Patch Embedding Module
2.2 Dual-Decoder for Intensity and Edge Reconstruction
3 Experiments and Results
3.1 Experimental Setup
3.2 Comparison with State-of-the-Art
3.3 Reconstruction Results
4 Conclusion
References
Self-supervised Learning of Morphological Representation for 3D EM Segments with Cluster-Instance Correlations
1 Introduction
2 Our Method
2.1 Instance-Level Contrastive Learning
2.2 Cluster-Instance Contrast Module
2.3 Data Augmentation
3 Experiments and Results
4 Conclusion
References
Calibrating Label Distribution for Class-Imbalanced Barely-Supervised Knee Segmentation
1 Introduction
2 Method
2.1 Cross Supervision for Semi-supervised Segmentation
2.2 Calibrating Label Distribution (CLD)
3 Experiments
4 Conclusion
References
Semi-supervised Medical Image Classification with Temporal Knowledge-Aware Regularization
1 Introduction
2 Methodology
2.1 Adaptive Pseudo Labeling (AdaPL)
2.2 Iterative Prototype Harmonizing (IPH)
2.3 Optimization
3 Experiments and Results
3.1 Dataset and Implementation Details
3.2 Comparison with State-of-the-Arts
3.3 Ablation Study
4 Conclusion
References
GaitForeMer: Self-supervised Pre-training of Transformers via Human Motion Forecasting for Few-Shot Gait Impairment Severity Estimation
1 Introduction
2 GaitForeMer: Gait Forecasting and Impairment Estimation TransforMer
2.1 Pre-training Procedure
2.2 Fine-Tuning Procedure
2.3 Baselines
3 Datasets
3.1 NTU RGB+D Dataset
3.2 MDS-UPDRS Dataset
4 Experiments
4.1 Using Motion Forecasting as an Effective Pre-training Task
4.2 Evaluating Fine-Tuning Strategies
4.3 Few-Shot Estimation of Gait Scores
4.4 Motion Forecasting Visualization
5 Conclusion
References
Semi-supervised Medical Image Segmentation Using Cross-Model Pseudo-Supervision with Shape Awareness and Local Context Constraints
1 Introduction
2 Methods
2.1 Cross-Model Pseudo-Supervision with Shape Awareness and Local Context Constraints
2.2 Loss Functions
3 Experiments and Results
4 Conclusion
References
MUSCLE: Multi-task Self-supervised Continual Learning to Pre-train Deep Models for X-Ray Images of Multiple Body Parts
1 Introduction
2 Methodologies
2.1 Multi-Dataset Momentum Contrastive Learning (MD-MoCo)
2.2 Multi-task Continual Learning
2.3 Fine-Tuning on Downstream Tasks
3 Experiment
4 Discussion and Conclusion
References
ShapePU: A New PU Learning Framework Regularized by Global Consistency for Scribble Supervised Cardiac Segmentation
1 Introduction
2 Method
2.1 EM Estimation of Multi-class Mixture Proportion
2.2 PU Learning with Marginal Probability
2.3 Global Consistency
3 Experiment
3.1 Experiment Setup
3.2 Ablation Study
3.3 Comparisons with Weakly Supervised Methods
3.4 Comparisons with PU Learning Methods
4 Conclusion
References
ProCo: Prototype-Aware Contrastive Learning for Long-Tailed Medical Image Classification
1 Introduction
2 Related Work
3 Methodology
3.1 Category Prototype and Adversarial Proto-instance
3.2 Prototype Recalibration
3.3 Proto-loss for Training
4 Experiments
4.1 Datasets and Evaluation
4.2 Implementation Details
4.3 Comparison with the State-of-the-Art
4.4 Ablation Study
5 Conclusion
References
Combining Mixed-Format Labels for AI-Based Pathology Detection Pipeline in a Large-Scale Knee MRI Study
1 Introduction
2 Data
3 Methods
3.1 Proposed Model Using Mixed-Format Labels
4 Results and Discussion
4.1 Performance of Proposed Model Using Mixed-Format Labels
5 Conclusions
References
Task-Oriented Self-supervised Learning for Anomaly Detection in Electroencephalography
1 Introduction
2 Methodology
2.1 Task-Oriented Self-supervised Learning
2.2 Anomaly Detection
3 Experiments and Results
4 Conclusion
References
Multiple Instance Learning with Mixed Supervision in Gleason Grading
1 Introduction
2 Method
2.1 Problem Fomulation
2.2 Instance Feature and Label Generation
2.3 Mixed Supervision Pipeline
3 Experiments
4 Conclusion
References
An Accurate Unsupervised Liver Lesion Detection Method Using Pseudo-lesions
1 Introduction
2 Method
2.1 Unsupervised Learning with Pseudo-lesions
2.2 U-Net Based Discriminator
2.3 Loss Function
2.4 Liver Lesion Detection by Image Gradient Perception
3 Experiments and Results
3.1 Datasets and Implementation
3.2 Comparison with State-of-the-Art Methods
3.3 Ablation Study
4 Conclusion
References
Addressing Class Imbalance in Semi-supervised Image Segmentation: A Study on Cardiac MRI
1 Introduction
2 Proposed Method
2.1 Basic Student-Teacher Framework
2.2 Dynamic Class-Aware Learning
3 Experiments and Results
3.1 Dataset and Implementation Details
3.2 Quantitative Performance Evaluation
3.3 Comparison with State-of-the-Art
3.4 Ablation Studies
4 Conclusion
References
Scribble2D5: Weakly-Supervised Volumetric Image Segmentation via Scribble Annotations
1 Introduction
2 Scribble2D5: Scribbles-Based Volumetric Segmentation
2.1 3D Pseudo Label Generation via Label Propagation
2.2 Scribble2D5 Network
3 Experiments
3.1 Datasets and Experimental Settings
3.2 Experimental Results
4 Conclusion and Discussion
References
Self-learning and One-Shot Learning Based Single-Slice Annotation for 3D Medical Image Segmentation
1 Introduction
2 Method
2.1 Problem Formulation
2.2 Pipeline
2.3 Feature Extraction Module
2.4 Screening Module
2.5 Reconstruction Module
2.6 Training Strategies
3 Experiments
3.1 Datesets and Metrics
3.2 Implementation
3.3 Compared Methods and Results
3.4 Ablation Study
4 Conclusions
References
Longitudinal Infant Functional Connectivity Prediction via Conditional Intensive Triplet Network
1 Introduction
2 Methods
2.1 Model Overview
2.2 Functional Connectivity Prediction Loss Design
2.3 Identity Conditional Module
3 Experiments
3.1 Dataset
3.2 Results and Visualization
3.3 Evaluation and Comparison
4 Conclusion
References
Leveraging Labeling Representations in Uncertainty-Based Semi-supervised Segmentation
1 Introduction
2 Method
2.1 Mean Teacher Formulation
2.2 Labeling Representation Prior
2.3 Uncertainty from a Labeling Representation
3 Results
4 Conclusion
References
Analyzing Brain Structural Connectivity as Continuous Random Functions
1 Introduction
2 Methodology
3 Experiments and Conclusions
References
Learning with Context Encoding for Single-Stage Cranial Bone Labeling and Landmark Localization
1 Introduction
2 Methods
2.1 Joint Cranial Bone Segmentation and Landmark Detection
2.2 Context Encoding
2.3 Regression of Spatial Landmark Configuration
3 Experiments and Results
3.1 Data Description
3.2 Implementation Details
3.3 Performance Evaluation
4 Discussion
5 Conclusion
References
Warm Start Active Learning with Proxy Labels and Selection via Semi-supervised Fine-Tuning
1 Introduction
2 Related Work
3 Proposed Method
3.1 Pre-ranking of Data via Pseudo Labels as a Proxy Task
3.2 Fully Supervised Training
3.3 Semi-supervised Training
3.4 Uncertainty Estimation and AL
4 Datasets and Experiments
4.1 Datasets
4.2 Experiments
5 Results
6 Conclusion
References
Intervention & Interaction Federated Abnormality Detection with Noisy Clients
1 Introduction
2 Intervention & Interaction Framework
2.1 Structural Causal Model for FADN
2.2 Intervention
2.3 Interaction
2.4 Learning Objective and Inference
3 Experiments
3.1 Dataset and Experimental Setup
3.2 Comparison with State-of-the-Arts
3.3 Further Analysis
4 Conclusion
References
SD-LayerNet: Semi-supervised Retinal Layer Segmentation in OCT Using Disentangled Representation with Anatomical Priors
1 Introduction
2 Model Architecture
2.1 Anatomical Factor Generation
2.2 Image Reconstruction
3 Experimental Results
4 Conclusion
References
Physiology-Based Simulation of the Retinal Vasculature Enables Annotation-Free Segmentation of OCT Angiographs
1 Introduction
2 Related Works
3 Methods
3.1 Physiology-Based Simulation of the Retinal Vasculature
3.2 Image Augmentations Based on the OCTA Acquisition Process
4 Experiments and Results
5 Discussion and Conclusion
References
Anomaly-Aware Multiple Instance Learning for Rare Anemia Disorder Classification
1 Introduction
2 Method
2.1 Instance-Level Feature Extraction
2.2 Anomaly-Aware Pooling
2.3 Optimization
3 Experiment
3.1 Classification
3.2 Explainability
3.3 Anomaly Recognition
3.4 CO2 Emission Related to Experiments
4 Conclusion
References
Unsupervised Domain Adaptation with Contrastive Learning for OCT Segmentation
1 Introduction
2 Methods
2.1 Simultaneously Learning from Labeled and Unlabeled Data
2.2 Pair Generation Strategy
2.3 Projection Heads to Extract Features for Image Segmentation
3 Experiments and Results
3.1 Evaluation on the Unlabeled Target Domain
3.2 Evaluation on the Labeled Source Domain
3.3 Ablations on Amount of Labeled Data
3.4 Segmentation Results Compared to Inter-grader Variability
4 Conclusions
References
Machine Learning – Model Interpretation
Neuro-RDM: An Explainable Neural Network Landscape of Reaction-Diffusion Model for Cognitive Task Recognition
1 Introduction
2 Method
2.1 Reaction-Diffusion Model for Functional Dynamics
2.2 Neuro-RDM: A Trainable and Better RDM by Linking PDE with GNN
3 Experiments
3.1 Result on Simulated Data
3.2 Application on Task-Based fMRI Data
4 Conclusion
References
Interpretable Graph Neural Networks for Connectome-Based Brain Disorder Analysis
1 Introduction
2 The Proposed Model
3 Experiments
4 Interpretation Analysis
5 Conclusion
References
Consistency-Preserving Visual Question Answering in Medical Imaging
1 Introduction
2 Method
3 Experiments and Results
4 Conclusions
References
Graph Emotion Decoding from Visually Evoked Neural Responses
1 Introduction
2 Methodology
2.1 Constructing an Emotion-Brain Bipartite Graph
2.2 Initializing Embeddings for Emotions and Brain Regions
2.3 Embedding Propagation Layers
3 Experiments
3.1 Dataset Description
3.2 Experimental Setup
3.3 Performance Comparison
3.4 Ablation Study
4 Conclusion and Future Work
References
Dual-Graph Learning Convolutional Networks for Interpretable Alzheimer's Disease Diagnosis
1 Introduction
2 Method
2.1 Graph Construction
2.2 Dual-Graph Learning
2.3 Graph Fusion and Objective Function
3 Experiments
3.1 Experimental Settings
3.2 Result Analysis
3.3 Ablation Analysis
3.4 Interpretability
3.5 Parameter Sensitivity Analysis
4 Conclusion
References
Asymmetry Disentanglement Network for Interpretable Acute Ischemic Stroke Infarct Segmentation in Non-contrast CT Scans
1 Introduction
2 Methodology
3 Experiments
4 Discussion and Conclusion
References
Regression Metric Loss: Learning a Semantic Representation Space for Medical Images
1 Introduction
2 Method
2.1 Learning a Semantically Interpretable Representation Space
2.2 Local Isometry: From Geodesic to Euclidean Distance
2.3 Hard Sample Pairs Mining
2.4 Making a Grounded Prediction
3 Experiments
3.1 Experimental Settings
3.2 Main Results
3.3 Ablation Studies
3.4 Data Efficiency
4 Conclusion
References
The (de)biasing Effect of GAN-Based Augmentation Methods on Skin Lesion Images
1 Introduction
2 Related Works
3 Experiments
3.1 Data and Training Details
3.2 Descriptive Statistics
3.3 Counterfactual Bias Insertion
4 Conclusions
References
Accurate and Explainable Image-Based Prediction Using a Lightweight Generative Model
1 Introduction
2 Method
3 Experiments
4 Results
5 Discussion
References
Interpretable Modeling and Reduction of Unknown Errors in Mechanistic Operators
1 Introduction
2 Method
2.1 Interpretable Modeling of Unknown Errors in Prior Physics
2.2 Inverse Estimation with Simultaneous Error Reduction
3 Experiments
3.1 Datasets
3.2 Generative Model Training
3.3 Evaluation of Generative Model
3.4 Evaluation of SOM
3.5 Optimization: Inverse Estimation
4 Conclusion
References
Sparse Interpretation of Graph Convolutional Networks for Multi-modal Diagnosis of Alzheimer's Disease
1 Introduction
2 Methods
2.1 Data Acquisition and Preprocessing
2.2 Brain Graph Construction
2.3 Sparse Interpretability of GCN
3 Results
4 Conclusions
References
Machine Learning – Uncertainty
FUSSNet: Fusing Two Sources of Uncertainty for Semi-supervised Medical Image Segmentation
1 Introduction
2 Method
2.1 Problem Definition
2.2 The FUSSNet Framework
2.3 Epistemic Uncertainty-Guided Training
2.4 Aleatoric Uncertainty-Guided Training
3 Experiments and Results
4 Conclusion
References
CRISP - Reliable Uncertainty Estimation for Medical Image Segmentation
1 Introduction
2 CRISP
3 Experimental Setup
3.1 Uncertainty Metrics
3.2 Data
3.3 Implementation Details
3.4 Experimental Setup
4 Results
5 Discussion and Conclusion
References
TBraTS: Trusted Brain Tumor Segmentation
1 Introduction
2 Method
2.1 Uncertainty and the Theory of Evidence for Medical Image Segmentation
2.2 Trusted Segmentation Network
2.3 Loss Function
3 Experiments
4 Conclusion
References
Layer Ensembles: A Single-Pass Uncertainty Estimation in Deep Learning for Segmentation
1 Introduction
2 Methodology
3 Materials and Implementation
4 Results
5 Conclusions
References
DEUE: Delta Ensemble Uncertainty Estimation for a More Robust Estimation of Ejection Fraction
1 Introduction
1.1 Clinical Background
1.2 Related Works
1.3 Our Contribution
2 Material and Methods
2.1 EF Dataset
2.2 Problem Definition and Method
3 Experiments and Results
4 Conclusion
References
Efficient Bayesian Uncertainty Estimation for nnU-Net
1 Introduction
2 Methods
2.1 Bayesian Inference
2.2 SGD Bayesian Inference
2.3 Multi-modal Posterior Sampling
3 Experiments
3.1 Experimental Setup
3.2 Results
4 Conclusion
References
Improving Trustworthiness of AI Disease Severity Rating in Medical Imaging with Ordinal Conformal Prediction Sets
1 Introduction
2 Methods
3 Experiments
3.1 Quantitative Experiments
3.2 Clinical Review of High Uncertainty Predictions
4 Related Work
5 Conclusion
References
Machine Learning Theory and Methodologies
Poisson2Sparse: Self-supervised Poisson Denoising from a Single Image
1 Introduction
2 Proposed Methodology: Poisson2Sparse
3 Experiments and Results
4 Conclusion
References
An Inclusive Task-Aware Framework for Radiology Report Generation
1 Introduction
2 Method
2.1 Task Distillation Module
2.2 Task-Aware Report Generation Module
2.3 Classification Token
2.4 Auto-Balance Mask Loss
3 Experiments and Results
3.1 Datasets and Metrics
3.2 Implementation Details
3.3 Performance Evaluation
3.4 Ablation Study
3.5 Qualitative Analysis
4 Conclusion
References
Removal of Confounders via Invariant Risk Minimization for Medical Diagnosis
1 Introduction
2 Removal of Confounders via Invariant Risk Minimization (ReConfirm)
3 Results and Discussion
4 Conclusion
References
A Self-guided Framework for Radiology Report Generation
1 Introduction
2 Related Works
3 Methods
3.1 Knowledge Distiller (KD)
3.2 Knowledge Matched Visual Extractor (KMVE)
3.3 Report Generator (RG)
4 Experiment
5 Results
6 Conclusion
References
D'ARTAGNAN: Counterfactual Video Generation
1 Introduction
2 Preliminaries
3 Method
4 Experimentation
5 Conclusion
References
TranSQ: Transformer-Based Semantic Query for Medical Report Generation
1 Introduction
2 Method
2.1 Visual Extractor
2.2 Semantic Encoder
2.3 Report Generator
3 Experiment
3.1 Datasets, Metrics and Settings
3.2 Results and Analyses
4 Conclusion and Discussion
References
Pseudo Bias-Balanced Learning for Debiased Chest X-Ray Classification
1 Introduction
2 Methodology
2.1 Problem Statement and Study Materials
2.2 Bias-Balanced Softmax
2.3 Bias Capturing with Generalized Cross Entropy Loss
2.4 Bias-Balanced Learning with Pseudo Bias
3 Experiments
4 Conclusion
References
Why Patient Data Cannot Be Easily Forgotten?
1 Introduction
2 Method
2.1 The Scrubbing Method
2.2 The Targeted Forgetting Method
3 Experiments
3.1 The Hardness of Patient-Wise Forgetting
3.2 Patient-Wise Forgetting Performance
4 Conclusion
References
Calibration of Medical Imaging Classification Systems with Weight Scaling
1 Introduction
2 Calibration Problem Formulation
3 Weight Scaling Based on the Predicted Confidence
4 Experimental Results
References
Online Reflective Learning for Robust Medical Image Segmentation
1 Introduction
2 Methodology
3 Experimental Results
4 Conclusion
References
Fine-Grained Correlation Loss for Regression
1 Introduction
2 Methodology
2.1 Effective PLC Loss to Optimize Linear Relationship
2.2 Coarse-to-Fine SRC to Regularize Rank
3 Experimental Results
3.1 Materials and Implementation Details
3.2 Quantitative and Qualitative Analysis
4 Conclusion
References
Suppressing Poisoning Attacks on Federated Learning for Medical Imaging
1 Introduction
2 Related Work
3 Proposed Method
4 Experiments
4.1 Datasets
4.2 FL Setup
4.3 Poisoning Attacks and Baseline Aggregation Rules
4.4 Results and Discussion
4.5 Conclusions and Future Work
References
The Intrinsic Manifolds of Radiological Images and Their Role in Deep Learning
1 Introduction
2 Methods: Intrinsic Dimension Estimation
3 Datasets and Tasks
4 Experiments and Results
4.1 The Intrinsic Dimension of Radiology Datasets
4.2 Generalization Ability, Learning Difficulty and Intrinsic Dimension
5 Conclusion
References
FedHarmony: Unlearning Scanner Bias with Distributed Data
1 Introduction
2 Methods
3 Implementation Details
4 Results and Discussion
5 Conclusion
References
Fast Unsupervised Brain Anomaly Detection and Segmentation with Diffusion Models
1 Introduction
2 Background
2.1 Compression Model
2.2 Denoising Diffusion Probabilistic Models
3 Proposed Anomaly Segmentation Method
4 Experiments
4.1 Anomaly Segmentation and Detection on Synthetic Anomalies
4.2 Anomaly Segmentation on MRI Data
4.3 Inference Time of Anomaly Segmentation on CT Data
5 Conclusions
References
Reliability of Quantification Estimates in MR Spectroscopy: CNNs vs Traditional Model Fitting
1 Introduction
2 Methods
2.1 Simulations
2.2 Quantification via Deep Learning
2.3 Quantification via Model Fitting
3 Results
4 Discussion
5 Conclusions
6 Data Availability Statement
References
AdaTriplet: Adaptive Gradient Triplet Loss with Automatic Margin Learning for Forensic Medical Image Matching
1 Introduction
2 Methods
2.1 Problem Statement
2.2 Triplet Loss
2.3 Adaptive Gradient Triplet Loss
2.4 AutoMargin: Adaptive Hard Negative Mining
3 Experiments
3.1 Datasets
3.2 Experimental Setup
3.3 Results
4 Discussion
References
Correction to: The Intrinsic Manifolds of Radiological Images and Their Role in Deep Learning
Correction to: Chapter “The Intrinsic Manifolds of Radiological Images and Their Role in Deep Learning” in: L. Wang et al. (Eds.): Medical Image Computing and Computer Assisted Intervention – MICCAI 2022, LNCS 13438, https://doi.org/10.1007/978-3-031-16452-1_65
Author Index