Interpretability of Machine Intelligence in Medical Image Computing

Name: Interpretability of Machine Intelligence in Medical Image Computing | 5th International Workshop, iMIMIC 2022, Held in Conjunction with MICCAI 2022, Singapore, Singapore, September 22, 2022, Proceedings
Brand: Springer
Price: 58.84 EUR
Availability: OnlineOnly

5th International Workshop, iMIMIC 2022, Held in Conjunction with MICCAI 2022, Singapore, Singapore, September 22, 2022, Proceedings

Mauricio Reyes Pedro Henriques Abreu Jaime Cardoso(Herausgeber*in)

Springer (Verlag)

Erschienen am 7. Oktober 2022

X, 125 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-031-17976-1 (ISBN)

58,84 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents
Interpretable Lung Cancer Diagnosis with Nodule Attribute Guidance and Online Model Debugging
1 Introduction
2 Materials
3 Methodology
3.1 Collaborative Model Architecture with Attribute-Guidance
3.2 Debugging Model with Semantic Interpretation
3.3 Explanation by Attribute-Based Nodule Retrieval
4 Experiments and Results
4.1 Implementation
4.2 Quantitative Evaluation
4.3 Trustworthiness Check and Interpretable Diagnosis
5 Conclusions
References
Do Pre-processing and Augmentation Help Explainability? A Multi-seed Analysis for Brain Age Estimation
1 Introduction
2 Related Work
3 Methods
4 Results
4.1 Performance
4.2 Voxel Agreement
4.3 Atlas-Based Analyses
4.4 Region Validation
5 Conclusion
References
Towards Self-explainable Transformers for Cell Classification in Flow Cytometry Data
1 Introduction
2 Related Work
3 Methods
3.1 Architecture
3.2 Preprocessing
3.3 Loss Function
3.4 Data Augmentation
4 Experiments
4.1 Data
4.2 Results
5 Conclusion
References
Reducing Annotation Need in Self-explanatory Models for Lung Nodule Diagnosis
1 Introduction
2 Method
3 Experimental Results
3.1 Prediction Performance of Nodule Attributes and Malignancy
3.2 Analysis of Extracted Features in Learned Space
3.3 Ablation Study
4 Conclusion
References
Attention-Based Interpretable Regression of Gene Expression in Histology
1 Introduction
2 Methods
2.1 Datasets
2.2 Multiple Instance Regression of Gene Expression
2.3 Attention-Based Model Interpretability
2.4 Evaluation of Performance and Interpretability
3 Experiments and Results
3.1 Network Training
3.2 Quantitative Model Evaluation
3.3 Attention-Based Identification of Hotspots and Patterns
3.4 Quantitative Evaluation of the Attention
4 Discussion
5 Conclusion
A Description of Selected Genes
B Detailed Model Evaluation
C Additional Visualizations
D Single-Cell Co-expression
References
Beyond Voxel Prediction Uncertainty: Identifying Brain Lesions You Can Trust
1 Introduction
2 Our Framework: Graph Modelization for Lesion Uncertainty Quantification
2.1 Monte Carlo Dropout Model and Voxel-Wise Uncertainty
2.2 Graph Dataset Generation
2.3 GCNN Architecture and Training
3 Material and Method
3.1 Data
3.2 Comparison with Known Approaches
3.3 Evaluation Setting
3.4 Implementation Details
4 Results and Discussion
5 Conclusion
References
Interpretable Vertebral Fracture Diagnosis
1 Introduction
1.1 Related Work
2 Methodology
2.1 Vertebral Fracture Detection
2.2 Semantic Concept Extraction (Correlation)
2.3 Visualization of Highly Correlating Concepts at Inference
3 Experimental Setup
4 Results and Discussion
4.1 Clinical Meaningfulness of Extracted Semantic Concepts
4.2 Single-Inference Concept Visualization
5 Conclusion
References
Multi-modal Volumetric Concept Activation to Explain Detection and Classification of Metastatic Prostate Cancer on PSMA-PET/CT
1 Introduction
2 Data
3 Method
3.1 Preprocessing
3.2 Detection
3.3 Classification
3.4 Explainable AI
4 Results
4.1 Detection
4.2 Classification
4.3 Explainable AI
5 Discussion
6 Conclusion
References
KAM - A Kernel Attention Module for Emotion Classification with EEG Data
1 Introduction
2 Related Work
3 Kernel Attention Module
4 Experiments
5 Conclusion
References
Explainable Artificial Intelligence for Breast Tumour Classification: Helpful or Harmful
1 Introduction
2 Related Work
2.1 XAI in Medicine
3 Model Setup
3.1 Data Pre-Processing
3.2 Model Architecture
4 Explanations
4.1 LIME
4.2 RISE
4.3 SHAP
5 Evaluating Explanations
5.1 One-Way ANOVA
5.2 Kendall's Tau
5.3 Radiologist Evaluation
5.4 Threats to Validity
6 Observations and Discussion
6.1 Discussion
A Appendix
A.1 Model Training Results
A.2 Choosing L Parameter for LIME
A.3 One-Way ANOVA Results
A.4 Pixel Agreement Statistics
A.5 Ranked Biased Overlap (RBO) Results
A.6 Kendall's Tau Results
A.7 Radiologist Opinions
A.8 Explanation Examples
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken