Interpretable and annotation-efficient learning for medical image computing : third International Workshop, iMIMIC 2020, Second International Workshop, MIL3iD 2020, and 5th International Workshop, LABELS 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4-8, 2020, Proceedings /: third International Workshop, iMIMIC 2020, Second International Workshop, MIL3iD 2020, and 5th International Workshop, LABELS 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4-8, 2020, Proceedings. (2020)

• Record Type: Book
• Other Titles: IMIMIC 2020; MIL3iD 2020; LABELS 2020
• Further Information: Note: Jaime Cardoso, Hien Van Nguyen, Nicholas Heller et al. (eds.).
• Other Names: Cardoso, Jaime S (Jaime dos Santos); Van Nguyen, Hien; Heller, Nicholas; iMIMIC (Workshop), 3rd; MIL3iD (Workshop), 2nd; LABELS (Workshop), 5th; International Conference on Medical Image Computing and Computer-Assisted Intervention, 23rd
• Contents: Intro -- Additional Workshop Editors -- iMIMIC 2020 Preface -- iMIMIC 2020 Organization -- MIL3iD 2020 Preface -- MIL3iD 2020 Organization -- LABELS 2020 Preface -- LABELS 2020 Organization -- Contents -- iMIMIC 2020 -- Assessing Attribution Maps for Explaining CNN-Based Vertebral Fracture Classifiers -- 1 Introduction -- 2 Methods -- 2.1 CNN Architectures -- 2.2 Prefix of Pre-trained U-Net Encoder -- 2.3 Custom CNN -- 2.4 Attribution Maps -- 2.5 Sanity Checks -- 3 Experiments -- 3.1 Vertebral Fracture Discrimination -- 3.2 Attribution Maps -- 4 Discussion and Conclusions -- References Projective Latent Interventions for Understanding and Fine-Tuning Classifiers -- 1 Introduction -- 2 Method -- 2.1 Parametric Embeddings -- 2.2 Projective Latent Constraints -- 2.3 Retraining the Classifier -- 3 Experiments -- 3.1 MNIST and CIFAR -- 3.2 Standard Plane Detection in Ultrasound Images -- 4 Discussion -- 5 Conclusion -- References -- Interpretable CNN Pruning for Preserving Scale-Covariant Features in Medical Imaging -- 1 Introduction -- 2 Methods -- 2.1 Notations -- 2.2 Representation of Scale Information -- 2.3 Bounding-Box Size vs. Image Size -- 2.4 Network Architectures and Tools 2.5 Datasets -- 3 Experiments and Results -- 3.1 Layer-Wise Quantification of Scale Invariance -- 3.2 Improvement of Transfer -- 4 Discussion -- 5 Conclusions and Future Work -- References -- Improving the Performance and Explainability of Mammogram Classifiers with Local Annotations -- 1 Introduction -- 2 Methods -- 2.1 The Extended Classifier -- 2.2 Perturbation-Based Explanations -- 2.3 Performance Measures -- 3 Experiments and Results -- 3.1 Datasets -- 3.2 Implementation Details -- 3.3 Classification Performance -- 3.4 Agreement with Local Annotations 3.5 Agreement with Perturbation-Based Explanations -- 4 Discussion -- References -- Improving Interpretability for Computer-Aided Diagnosis Tools on Whole Slide Imaging with Multiple Instance Learning and Gradient-Based Explanations -- 1 Introduction -- 2 Related Work and Motivations -- 3 Proposed Methods -- 4 Experiments and Results -- 5 Conclusion -- References -- Explainable Disease Classification via Weakly-Supervised Segmentation -- 1 Introduction -- 2 Method -- 2.1 Dataset(s) -- 2.2 Model -- 2.3 Training -- 2.4 Deriving the Heatmaps -- 3 Experiments -- 4 Results 5 Conclusion and Discussion -- References -- Reliable Saliency Maps for Weakly-Supervised Localization of Disease Patterns -- 1 Introduction -- 2 Methods -- 2.1 Dataset, Preprocessing, Training Parameters -- 2.2 Radiologist Survey -- 3 Experiments -- 3.1 Results -- 4 Discussion and Conclusion -- References -- Explainability for Regression CNN in Fetal Head Circumference Estimation from Ultrasound Images -- 1 Introduction -- 2 Saliency Map Methods for Regression CNN -- 2.1 State-of-the-Art Saliency Maps in CNN -- 2.2 Evaluation of Explanation Methods Based on Perturbation -- 3 Experiments
• Publisher Details: Cham : Springer
• Publication Date: 2020
• Extent: 1 online resource (305 p.)
• Subjects: 616.07/57; Diagnostic imaging -- Data processing -- Congresses; Computer-assisted surgery -- Congresses; Application software; Artificial intelligence; Bioinformatics; Optical data processing; Pattern perception; Electronic books; Electronic books
• Languages: English
• ISBNs: 9783030611668; 3030611663
• Related ISBNs: 9783030611651; 3030611655
• Notes: Note: Online resource; title from PDF title page (SpringerLink, viewed December 2, 2020).
• Access Rights: Legal Deposit; Only available on premises controlled by the deposit library and to one user at any one time; The Legal Deposit Libraries (Non-Print Works) Regulations (UK).
• Access Usage: Restricted: Printing from this resource is governed by The Legal Deposit Libraries (Non-Print Works) Regulations (UK) and UK copyright law currently in force.