Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 Workshops (Lecture Notes in Computer Science)

✍ Scribed by Jonghye Woo; Alessa Hering; Silva Wilson; Xiang Li; Huazhu Fu; Xiaofeng Liu; Fangxu Xing; Sanjay Purushotham; Tejas S. Mathai; Pritam Mukherjee; Max De Grauw; Regina Beets Tan; Valentina Corbetta; Elmar Kotter; Mauricio Reyes; Christian F. Baumgartner; Quanzheng Li; Richard Leahy; Bin Dong; Hao Chen; Yuankai Huo; Jinglei LV; Xinxing Xu; Dwarikanath Mahapatra; Li Cheng; Caroline Petitjean; Benoît Presles

• Tongue: English
• Leaves: 394
• Category: Library

✦ Table of Contents

Workshop Editors
MTSAIL 2023 Preface
MTSAIL 2023 Organization
LEAF 2023 Preface
LEAF 2023 Organization
AI4Treat 2023 Preface
AI4Treat 2023 Organization
MMMI 2023 Preface
MMMI 2023 Organization
REMIA 2023 Preface
REMIA 2023 Organization
Contents
Proceedings of the First MICCAI Workshop on Time-Series Data Analytics and Learning (MTSAIL 2023)
Learning Dynamic MRI Reconstruction with Convolutional Network Assisted Reconstruction Swin Transformer
1 Introduction
2 Materials and Methods
2.1 SADXNet Assisted Reconstruction Swin Transformer
2.2 Data Cohort
3 Results
4 Discussion
5 Conclusion
References
A Groupwise Method for the Reconstruction of Hypergraph Representation of Resting-State Functional Networks
1 Introduction
2 Method
2.1 Hypergraph
2.2 Hypergraph Construction
2.3 Individual Hypergraph Incident Matrix
3 Results
3.1 Datasets
3.2 Competing Methods
3.3 Classification Results
3.4 Statistical Significance of Results
4 Conclusion
References
MomentaMorph: Unsupervised Spatial-Temporal Registration with Momenta, Shooting, and Correction
1 Introduction
2 Method
2.1 Momenta, Shooting, Correction
2.2 Learning
3 Experiments
3.1 Synthetic Elastic 2D Data
3.2 Tongue 3D tMRI Data
4 Conclusion and Discussion
References
FusionNet: A Frame Interpolation Network for 4D Heart Models
1 Introduction
2 Dataset
3 FusionNet Architecture
4 Experiment and Discussion
4.1 Comparison with Existing Methods
4.2 Ablation Study
4.3 Robustness to Changes in Frame Intervals
5 Conclusion
References
A New Large-Scale Video Dataset of the Eyelid Opening Degree for Deep Regression-Based PERCLOS Estimation
1 Introduction
2 Proposed Dataset
2.1 Experimental Procedure
2.2 Time Series Data Analysis
3 Proposed Approach
4 Experiments and Analysis
4.1 Implementation Details
4.2 Experimental Results
5 Conclusions and Future Work
References
Proceedings of the First MICCAI Workshop on Lesion Evaluation and Assessment with Follow-up (LEAF 2023)
A Hierarchical Descriptor Framework for On-the-Fly Anatomical Location Matching Between Longitudinal Studies
1 Introduction
2 Method
2.1 Descriptor Computation
2.2 Similarity
2.3 Hierarchical Search
3 Experiments
3.1 Comparison Study
3.2 Mixed Modality Dataset
4 Conclusion
References
A Two-Species Model for Abnormal Tau Dynamics in Alzheimer's Disease
1 Introduction
2 Methodology
3 Results
4 Conclusions
References
Outlier Robust Disease Classification via Stochastic Confidence Network
1 Introduction
2 Methods
2.1 Decoded Latent Matrix (DLM)
2.2 Stochastic Loss Functions
2.3 Resample Outliers
3 Experiments
3.1 Ablation Study
3.2 Quantitative Analysis
3.3 Qualitative Analysis
4 Conclusion
References
Efficient Registration of Longitudinal Studies for Follow-Up Lesion Assessment by Exploiting Redundancy and Composition of Deformations
1 Introduction
2 Methods
3 Experiments
4 Results
5 Discussion and Conclusion
References
Proceedings of the AI for Treatment Response Assessment and predicTion Workshop (AI4Treat 2023)
Graph-Based Multimodal Multi-lesion DLBCL Treatment Response Prediction from PET Images
1 Introduction
2 Related Work
3 Method
4 Experiments
5 Results
6 Conclusion
References
RPTK: The Role of Feature Computation on Prediction Performance
1 Introduction
2 Material and Methods
3 Results
4 Discussion
References
Proceedings of the Fourth International Workshop on Multiscale Multimodal Medical Imaging (MMMI 2023)
M2Fusion: Bayesian-Based Multimodal Multi-level Fusion on Colorectal Cancer Microsatellite Instability Prediction
1 Introduction
2 Method
2.1 Bayesian-Based Multi-modality Fusion Model
2.2 MSI Prediction on Single Modality
2.3 Model Prediction Fusion on Multiple Levels
3 Experiments
3.1 Dataset
3.2 Experimental Design
4 Result
5 Conclusion
References
Query Re-Training for Modality-Gnostic Incomplete Multi-modal Brain Tumor Segmentation
1 Introduction
2 Method
2.1 Architecture Overview
2.2 Modality-Gnostic Transformer Module
2.3 Query Re-Training Strategy
2.4 Loss Function
3 Experiments
3.1 Implementation Details
3.2 Performance Comparison
3.3 Ablation Study
4 Conclusion
References
MAD: Modality Agnostic Distance Measure for Image Registration
1 Introduction
2 Related Works
3 Methods
4 Experiments and Discussion
4.1 Experiment 1: Loss Landscapes
4.2 Experiment 2: Recover Random Transformations
4.3 Experiment 3: Ablation Study
5 Conclusion
References
Multimodal Context-Aware Detection of Glioma Biomarkers Using MRI and WSI
1 Introduction
2 Proposed Methodology
3 Experimental Details
3.1 Dataset
3.2 Data Pre-processing
3.3 Training Details
3.4 Evaluation Metrics
4 Results and Discussion
5 Conclusions
References
Modality Cycles with Masked Conditional Diffusion for Unsupervised Anomaly Segmentation in MRI
1 Introduction
2 Multi-modal Cycles with Masked Conditional Diffusion
3 Experiments
4 Conclusion
References
BreastRegNet: A Deep Learning Framework for Registration of Breast Faxitron and Histopathology Images
1 Introduction
2 Methods
2.1 Data Description and Analysis
2.2 Registration Network
2.3 Training
2.4 Evaluation Metrics
2.5 Implementation Details
3 Results
4 Discussion and Conclusion
References
Osteoarthritis Diagnosis Integrating Whole Joint Radiomics and Clinical Features for Robust Learning Models Using Biological Privileged Information
1 Introduction
2 Methods
2.1 Dataset
2.2 Statistical and Machine Learning Approaches
3 Results
3.1 LUPI and Non-LUPI Models
3.2 Feature Integration Comparison
3.3 Feature Occurrence and Importance
4 Discussion
5 Conclusion
References
Graph-Based Counterfactual Causal Inference Modeling for Neuroimaging Analysis
1 Introduction
2 Related Work
2.1 Counterfactual Outcome Estimation
2.2 Continuous Treatment Effect Estimation
2.3 Traditional Correlation-Based PET Image Analysis Methods
3 Methodology
3.1 Problem Setting
3.2 VCNet
3.3 Graph-VCNet
4 Experiment
4.1 Dataset
4.2 Experiment Setting
4.3 Prediction Performance
5 Conclusion and Discussion
References
Synthesising Brain Iron Maps from Quantitative Magnetic Resonance Images Using Interpretable Generative Adversarial Networks
1 Introduction
2 Related Work
2.1 Supervised Image-to-Image Translation
2.2 Interpretable Deep Learning
3 Methods
3.1 Data
3.2 Image Registration
3.3 Model Architecture
3.4 Model Loss
3.5 Interpretability
3.6 Performance Metrics
4 Experiments and Results
4.1 Implementation Details
4.2 Analysis of Model Architecture
4.3 Analysis of Model Loss
4.4 Analysis of Interpretability
5 Discussion
References
Identifying Shared Neuroanatomic Architecture Between Cognitive Traits Through Multiscale Morphometric Correlation Analysis
1 Introduction
2 Methods
2.1 Bivariate Linear Mixed Effect Model
2.2 Efficient Average Information REML Algorithm
3 Experimental Results
3.1 Materials
3.2 Simulation Results
3.3 Whole Brain Morphometric Correlation and Morphometricity
3.4 Brain ROI-Based Morphometric Correlation and Morphometricity
4 Conclusion and Discussions
References
Noisy-Consistent Pseudo Labeling Model for Semi-supervised Skin Lesion Classification
1 Introduction
2 Method
2.1 Overview
2.2 Problem Definition
2.3 Noisy-Consistent Sample Learning
2.4 Attentive Clustered Feature Integration
3 Experiments
3.1 Dataset and Implementation
3.2 Results
4 Discussion and Conclusion
References
Hessian-Based Similarity Metric for Multimodal Medical Image Registration
1 Introduction
2 Method
2.1 Defining the Hessian-Based Similarity Metric
2.2 Transforming Hessians
2.3 Implementing the Metric in an Affine Registration Scheme
3 Experiments
3.1 Robustness to Intensity Nonuniformities
3.2 Quantitative Results
4 Discussion and Conclusion
References
Hybrid Multimodality Fusion with Cross-Domain Knowledge Transfer to Forecast Progression Trajectories in Cognitive Decline
1 Introduction
2 Materials and Methodology
2.1 Subjects and Image Preprocessing
2.2 Proposed Method
3 Experiment
4 Conclusion
References
MuST: Multimodal Spatiotemporal Graph-Transformer for Hospital Readmission Prediction
1 Introduction
2 Methodology
2.1 Problem Formulation
2.2 Feature Extraction for Single Modality
2.3 Multimodal Fusion
3 Experiments
4 Conclusion
References
Groupwise Image Registration with Atlas of Multiple Resolutions Refined at Test Phase
1 Introduction
2 Method
2.1 Architecture of SETGen
2.2 Test-Time Atlas Generation
3 Experiments
3.1 Datasets
3.2 Baselines
3.3 Evaluation Metrics
3.4 Results of Adaptation with Original Resolution
3.5 Results on Multi-scales Atlas Building
3.6 Ablation Study
4 Conclusion
References
Anatomy-Aware Lymph Node Detection in Chest CT Using Implicit Station Stratification
1 Introduction
2 Method
3 Experiment
4 Conclusion
References
Leveraging Contrastive Learning with SimSiam for the Classification of Primary and Secondary Liver Cancers
1 Introduction
2 Methods
2.1 Data Description and Preprocessing
2.2 Pre-training Models Using SimSiam
3 Experimental Results and Discussion
4 Conclusion
References
Proceeding of the Second International Workshop on Resource-Efficient Medical Image Analysis (REMIA 2023)
Operating Critical Machine Learning Models in Resource Constrained Regimes
1 Introduction
2 Methods for Resource Efficiency
3 Data and Experiments
4 Discussion
5 Conclusion
References
Data Efficiency of Segment Anything Model for Optic Disc and Cup Segmentation
1 Background
2 Methods
2.1 Datasets
2.2 Experiments
3 Results
3.1 OD Segmentation
3.2 OC Segmentation
4 Discussion and Conclusions
References
Anisotropic Hybrid Networks for Liver Tumor Segmentation with Uncertainty Quantification
1 Introduction
2 Materials and Methods
2.1 Dataset
2.2 Segmentation Pipelines
2.3 Pipeline Multi-class
2.4 Dual Binary Pipeline
2.5 Post-processing Module
2.6 Lesion Uncertainty Quantification
2.7 Evaluation Protocol
2.8 Ablation Study
2.9 Implementation Details
3 Results and Discussion
4 Conclusion
References
PLMVQA: Applying Pseudo Labels for Medical Visual Question Answering with Limited Data
1 Introduction
2 Related Work
3 Methodology
3.1 Model Architecture
3.2 Constructing Pseudo Labels from Limited Data
3.3 Applying Pseudo Labels for Medical VQA
4 Experiments
4.1 Dataset and Evaluation Metric
4.2 Experimental Details
4.3 Results
4.4 Ablation Study
4.5 Qualitative Analysis
5 Conclusion
References
SAM-U: Multi-box Prompts Triggered Uncertainty Estimation for Reliable SAM in Medical Image
1 Introduction
2 Method
2.1 Mask Selection Strategy
2.2 SAM with Multi-box Prompts
2.3 Uncertainty Estimation of SAM with Multi-box Prompts
3 Experiments and Results
3.1 Data and Implementation Details
3.2 Evaluation Protocols
3.3 Quantitative Evaluation
3.4 Qualitative Comparison
4 Discussion and Conclusion
References
Author Index