Artificial Intelligence in Medicine - 21st International Conference on Artificial Intelligence in Medicine, AIME 2023, Portorož, Slovenia, June 12–15, 2023, Proceedings 9783031343438, 9783031343445

Table of contents :
Preface
Organization
Contents
Machine Learning and Deep Learning
Survival Hierarchical Agglomerative Clustering: A Semi-Supervised Clustering Method Incorporating Survival Data
1 Introduction
2 Materials and Methods
2.1 Hierarchical Survival Agglomeration
2.2 Benchmark Experiments
2.3 Case Study: Atrial Fibrillation Phenotypes
3 Results and Discussion
3.1 Benchmark Experiments
3.2 Case Study: Atrial Fibrillation Phenotypes
3.3 Limitations
4 Conclusions
5 Implementation Details and Code Availability
References
Boosted Random Forests for Predicting Treatment Failure of Chemotherapy Regimens
1 Introduction
2 Real World Evidence and Oncology Data Overview
3 Feature Engineering
4 An Empirical Design Exploration Framework
5 Conclusion
References
A Binning Approach for Predicting Long-Term Prognosis in Multiple Sclerosis
1 Introduction
2 Dataset
3 Methodology
3.1 Defining the Target
3.2 Machine Learning Model
3.3 Evaluation Procedure
4 Results and Discussion
5 Conclusion
References
Decision Tree Approaches to Select High Risk Patients for Lung Cancer Screening Based on the UK Primary Care Data
1 Introduction
2 Data and Methods
3 Results
4 Conclusion
References
Causal Discovery with Missing Data in a Multicentric Clinical Study
1 Introduction
2 Background
3 Multicentric Clinical Data on Endometrial Cancer
4 Experiments
5 Discussion and Conclusions
References
Novel Approach for Phenotyping Based on Diverse Top-K Subgroup Lists
1 Introduction
2 Problem Definition and Background
3 DSLM Algorithm
4 Experiments and Discussion
5 Conclusions
References
Patient Event Sequences for Predicting Hospitalization Length of Stay
1 Introduction
2 Transformer Models for EHR
3 Empirical Evaluation and Results
3.1 Data and Experimental Setting
3.2 Results
4 Conclusion
References
Autoencoder-Based Prediction of ICU Clinical Codes
1 Introduction
2 Methods
3 Results
4 Discussion
References
Explainability and Transfer Learning
Hospital Length of Stay Prediction Based on Multi-modal Data Towards Trustworthy Human-AI Collaboration in Radiomics
1 Introduction
2 Predicting Hospital Length of Stay Using X-ray Images
3 Explaining Length of Stay Predictions to Humans
4 Discussion
References
Explainable Artificial Intelligence for Cytological Image Analysis
1 Introduction
2 Background
2.1 Differential Blood Count Using Quantitative Phase Imaging
2.2 Image Acquisition and Data Processing Pipeline
2.3 Evaluating Explainability and Interpretability
3 XAI Dashboard Prototype
3.1 Module 1: General Information on Training and Validation
3.2 Module 2: Image Samples of Classified Cells
3.3 Module 3: Morphological Features in a Scatter Plot
3.4 Module 4: Revealing Relevant Areas of an Image Using LIME
4 User Study Results
4.1 Evaluation of the Overall XAI Dashboard
4.2 Comparison of the XAI Modules
5 Discussion and Conclusion
References
Federated Learning to Improve Counterfactual Explanations for Sepsis Treatment Prediction
1 Introduction
2 Related Work
3 Methodology
3.1 Data Acquisition and Preparation
3.2 Sepsis Treatment Prediction
3.3 Counterfactual Explanation Generation
3.4 Counterfactual Explanation Evaluation
4 Experiments
4.1 Counterfactual Explanation Quality Comparison
4.2 Effects of Data Imbalance on Counterfactual Quality
4.3 Clinical Application
5 Conclusion, Limitations, and Future Work
References
Explainable AI for Medical Event Prediction for Heart Failure Patients
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Database and Machine Learning Model
3.2 Global and Local Explanations
3.3 Model Explanation Pipeline
4 Results and Discussion
4.1 Possible Implementation in CDSS
5 Conclusion and Future Work
References
Adversarial Robustness and Feature Impact Analysis for Driver Drowsiness Detection
1 Introduction
2 Analysis Methodology
3 Results and Discussion
4 Conclusions
References
Computational Evaluation of Model-Agnostic Explainable AI Using Local Feature Importance in Healthcare
1 Introduction
2 Method
3 Results
4 Conclusion
References
Batch Integrated Gradients: Explanations for Temporal Electronic Health Records
1 Introduction
2 Method
3 Experiment
4 Conclusion
References
Improving Stroke Trace Classification Explainability Through Counterexamples
1 Introduction
2 Trace Classification Explainability and the Role of Counterexamples
3 Results
4 Conclusions
References
Spatial Knowledge Transfer with Deep Adaptation Network for Predicting Hospital Readmission
1 Introduction
2 Deep Adaptation Network
3 The Proposed ERR-TDAN Framework
3.1 Representation Learning of Temporal EHR Data with LSTM
3.2 Learning Transferable Features and Predictions
3.3 Model Optimization via a Customized Loss Function
4 Data
5 Experimental Setup and Results
5.1 Can We Enhance Readmission Risk Prediction for a Target Hospital by Utilizing Data from Another Hospital?
5.2 What is the Retrospective Optimal Amount of EHR Data Needed for Future Predictions?
5.3 How Often Do We Need to Retrain the Model to Achieve Optimal Performance?
6 Discussion and Conclusion
References
Dealing with Data Scarcity in Rare Diseases: Dynamic Bayesian Networks and Transfer Learning to Develop Prognostic Models of Amyotrophic Lateral Sclerosis
1 Introduction
2 Study Population and Data Preprocessing
2.1 Data Preprocessing
3 Methods
3.1 Data Simulation via DBNs
3.2 Model Development Strategy
4 Results
4.1 Data Simulation via DBNs
4.2 Model Performance
5 Discussion and Conclusions
References
Natural Language Processing
A Rule-Free Approach for Cardiological Registry Filling from Italian Clinical Notes with Question Answering Transformers
1 Introduction
1.1 Research Goal
1.2 Related Work
2 Materials and Methods
2.1 Dataset
2.2 Transformer-Based Models
2.3 Experimental Design
3 Results
4 Discussion
5 Conclusion
References
Classification of Fall Types in Parkinson's Disease from Self-report Data Using Natural Language Processing
1 Introduction
2 Methods
2.1 Fall Self-report Dataset
2.2 Data Pre-processing
2.3 Fall Class Classification
3 Results
3.1 Demographic Information
3.2 Binary Classification Hyperparameter Selection
3.3 Binary Classifier Performance
3.4 Ablation Study of Features in the Ensemble Model
3.5 Binary Classification Using RoBERTa
4 Discussion
4.1 Better Performance from ML Classifiers
4.2 A Lack of Gender Differences in Fall Frequency and Type
4.3 Future Work
5 Conclusion
References
BERT for Complex Systematic Review Screening to Support the Future of Medical Research
1 Introduction and Motivations
2 Related Work
3 Methodology
3.1 Dataset Creation
3.2 Data Augmentation
3.3 BERT for Text Classification
4 Experimental Results
4.1 Dataset
4.2 Implementation Details
4.3 Data Augmentation
4.4 Model Performance
5 Conclusions
References
GGTWEAK: Gene Tagging with Weak Supervision for German Clinical Text
1 Introduction
2 Related Work
3 Methods
3.1 Dataset and Annotation
3.2 Labelling Function Development
3.3 Labelling Function Aggregation
3.4 Named Entity Recognition Models
4 Results
4.1 Labelling Function Analysis
4.2 Evaluation Against Gold-Standard Annotations
5 Discussion and Limitations
6 Conclusion and Future Work
References
Soft-Prompt Tuning to Predict Lung Cancer Using Primary Care Free-Text Dutch Medical Notes
1 Introduction
2 Methodology
3 Results and Discussion
4 Conclusions
References
Machine Learning Models for Automatic Gene Ontology Annotation of Biological Texts
1 Introduction
2 Methods
3 Results and Discussion
4 Conclusions
References
Image Analysis and Signal Analysis
A Robust BKSVD Method for Blind Color Deconvolution and Blood Detection on H&E Histological Images
1 Introduction
1.1 Related Work
2 Material and Methods
2.1 Bayesian k-SVD for Blind Color Deconvolution
2.2 Robust Blind Color Deconvolution and Blood Detection
2.3 Databases
3 Experiments
3.1 Blood Detection
3.2 Effect of Blood in the Estimation of the Color-Vector Matrix
4 Conclusions
References
Can Knowledge Transfer Techniques Compensate for the Limited Myocardial Infarction Data by Leveraging Hæmodynamics? An in silico Study
1 Introduction
2 Dataset Generation
3 Deep Learning Models for MI Risk Prediction
3.1 Scenario 1 – Transfer Learning
3.2 Scenario 2 – Multitask Learning
4 Conclusions
References
COVID-19 Diagnosis in 3D Chest CT Scans with Attention-Based Models
1 Introduction
2 Related Works
2.1 COVID-19 Datasets
2.2 Convolutional Neural Networks for COVID-19 Detection
2.3 Attention-Based Models for COVID-19 Detection
3 Methodology
3.1 Dataset
3.2 TimeSformer
3.3 Original Attention Schemes
3.4 New Attention Schemes
3.5 Experimental Setup
3.6 Evaluation Metrics
4 Results and Discussion
5 Conclusion
References
Generalized Deep Learning-Based Proximal Gradient Descent for MR Reconstruction
1 Introduction
2 Method - Proximal Network as a Regularization Term
3 Experiments and Results
4 Discussion and Conclusion
References
Crowdsourcing Segmentation of Histopathological Images Using Annotations Provided by Medical Students
1 Introduction
2 Proposed Methodology
3 Experiments
3.1 Experimental Setup
3.2 Results
4 Conclusion
References
Automatic Sleep Stage Classification on EEG Signals Using Time-Frequency Representation
1 Introduction
2 State of the Art
3 Method
3.1 Dataset Preprocessing
3.2 Contextual Information
3.3 Finetuning a Deep Network to Multi-electrode Data
4 Experiments
4.1 Dataset Used
4.2 Folds and Metrics
4.3 Time-Frequency Representations
4.4 Increasing Context Information
4.5 Removing Spatial Information
4.6 Comparison with State-of-the-Art Methods
5 Discussion
6 Conclusion
References
Learning EKG Diagnostic Models with Hierarchical Class Label Dependencies
1 Introduction
2 Related Work
3 Methodology
3.1 Low-dimensional EKG Representation
3.2 Multi-label Classification Model
3.3 Training Models
3.4 Making Predictions
4 Experiments
4.1 Data
4.2 Methods
4.3 Metrics
4.4 Results
5 Conclusion
References
Discriminant Audio Properties in Deep Learning Based Respiratory Insufficiency Detection in Brazilian Portuguese
1 Introduction
2 Related Work
3 Methodology
4 Results and Discussion
5 Conclusion and Future Work
References
ECGAN: Self-supervised Generative Adversarial Network for Electrocardiography
1 Introduction
2 The ECGAN Model
3 Experimental Analysis
4 Discussion
References
Data Analysis and Statistical Models
Nation-Wide ePrescription Data Reveals Landscape of Physicians and Their Drug Prescribing Patterns in Slovenia
1 Introduction
2 Data and Methods
2.1 Data
2.2 Methods
3 Results and Discussion
3.1 Non-drug Prescription Data Aids in the Identification of Physician Specializations
3.2 Investigating Previously-Identified, Specialization-Specific Drugs
4 Conclusion and Future Directions
References
Machine Learning Based Prediction of Incident Cases of Crohn's Disease Using Electronic Health Records from a Large Integrated Health System*-12pt
1 Introduction
2 Methods
2.1 Data and Study Population
2.2 Phenotyping Algorithm
2.3 Risk Prediction Problem Framing
2.4 Data Pre-processing and Feature Extraction
2.5 Predictive Modeling and Evaluation
3 Results
4 Discussion
4.1 Clinical and Technical Significance
4.2 Limitations and Future Work
5 Conclusion
References
Prognostic Prediction of Pediatric DHF in Two Hospitals in Thailand
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Data
3.2 Models and Settings
3.3 Generalizability
4 Results
5 Discussion
6 Conclusions and Future Work
References
The Impact of Bias on Drift Detection in AI Health Software
1 Introduction
2 Methods
2.1 Dataset
2.2 Methodology
3 Results
3.1 Protected Attribute: Ethnicity
3.2 Protected Attribute: Gender
4 Conclusions
References
A Topological Data Analysis Framework for Computational Phenotyping
1 Introduction
2 Materials and Methods
2.1 Dataset and Experimental Setup
3 Results
4 Conclusion and Future Works
References
Ranking of Survival-Related Gene Sets Through Integration of Single-Sample Gene Set Enrichment and Survival Analysis
1 Introduction
2 Methods
2.1 Data
2.2 Single-Sample Gene Set Enrichment Analysis
2.3 Gene Set Ranking for Survival Analysis
2.4 Robustness Estimation
3 Results
4 Discussion
4.1 Analysis of the Method's Robustness
4.2 Varying Splitting Threshold
5 Conclusions
References
Knowledge Representation and Decision Support
Supporting the Prediction of AKI Evolution Through Interval-Based Approximate Predictive Functional Dependencies
1 Introduction
2 Predicting the Evolution of Pathological States
2.1 Extending APFDs to Deal with Interval-Based Predictions
3 Experiments: Dataset and Data Transformation
3.1 Results
4 Conclusions
References
Integrating Ontological Knowledge with Probability Data to Aid Diagnosis in Radiology
1 Introduction
1.1 Ontologies
1.2 Radiology Gamuts Ontology
2 Methods
2.1 Ethics Statement
2.2 Patient Cohort
2.3 Occurrence and Co-occurrence of Entities
3 Results
3.1 Web Interface
3.2 Bayesian Network
4 Discussion
References
Ontology Model for Supporting Process Mining on Healthcare-Related Data
1 Introduction
2 Process Mining on Electronic Health Records
3 Clinical Process Model Ontology
4 Application of the CPMO in Prostate Cancer Use Case
5 Conclusion and Future Work
References
Real-World Evidence Inclusion in Guideline-Based Clinical Decision Support Systems: Breast Cancer Use Case
1 Introduction
2 Methodology
3 Breast Cancer Use Case
4 Conclusions and Future Work
References
Decentralized Web-Based Clinical Decision Support Using Semantic GLEAN Workflows
1 Introduction
2 Methods
2.1 Decentralized Collaborating on Multimorbidity Patients
2.2 Conflict Detection Between Multimorbidity Workflows
3 Discussion and Conclusion
References
An Interactive Dashboard for Patient Monitoring and Management: A Support Tool to the Continuity of Care Centre
1 Introduction
2 Dashboard Implementation
2.1 Data Integration and Preprocessing
2.2 Intensity of Medical Care
2.3 Conformance Checking
3 Interface Design and Functionality
4 Conclusions and Future Work
References
A General-Purpose AI Assistant Embedded in an Open-Source Radiology Information System
1 Introduction
2 LibreHealth RIS Architecture
2.1 LibreHealth Radiology Module
2.2 LibreHealth Radiology Artificial Intelligence Model Service
3 Methods
3.1 OHIF Viewer and Its Extensions
3.2 AI Model Service
3.3 Swarm Learning and Few Shot Learning
4 Discussion
5 Conclusion
References
Management of Patient and Physician Preferences and Explanations for Participatory Evaluation of Treatment with an Ethical Seal
1 Introduction
2 Background
3 Method
4 Results
5 Conclusions
References
Author Index

Polecaj historie

Artificial Intelligence in Medicine: 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Minneapolis, MN, USA, August 25–28, 2020, Proceedings [1st ed.] 9783030591366, 9783030591373

The LNAI 12299 constitutes the papers of the 18th International Conference on Artificial Intelligence in Medicine, AIME

732 99 11MB Read more

Modeling Decisions for Artificial Intelligence. 20th International Conference, MDAI 2023 Umeå, Sweden, June 19–22, 2023 Proceedings 9783031334979, 9783031334986

447 73 9MB Read more

Modeling Decisions for Artificial Intelligence: 20th International Conference, MDAI 2023, Umeå, Sweden, June 19–22, 2023, Proceedings 9783031334986, 9783031334979, 3031334981

441 44 24MB Read more

Modeling Decisions for Artificial Intelligence: 20th International Conference, MDAI 2023, Umeå, Sweden, June 19–22, 2023, Proceedings 3031334973, 9783031334979

This book constitutes the refereed proceedings of the 20th International Conference on Modeling Decisions for Artificial

505 103 9MB Read more

Artificial General Intelligence: 16th International Conference, AGI 2023, Stockholm, Sweden, June 16–19, 2023, Proceedings 303133468X, 9783031334689

This book constitutes the refereed proceedings of the 16th International Conference on Artificial General Intelligence,

643 142 24MB Read more

Artificial General Intelligence: 16th International Conference, AGI 2023, Stockholm, Sweden, June 16–19, 2023, Proceedings 9783031334696, 9783031334689, 3031334698

This book constitutes the refereed proceedings of the 16th International Conference on Artificial General Intelligence,

315 21 30MB Read more

Artificial General Intelligence. 16th International Conference, AGI 2023 Stockholm, Sweden, June 16–19, 2023 Proceedings 9783031334689, 9783031334696

465 52 22MB Read more

Artificial Intelligence in Medicine: 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Poznan, Poland, June 26–29, 2019, Proceedings [1st ed.] 978-3-030-21641-2;978-3-030-21642-9

This book constitutes the refereed proceedings of the 17th Conference on Artificial Intelligence in Medicine, AIME 2019,

709 97 33MB Read more

Artificial Intelligence in Education: 24th International Conference, AIED 2023, Tokyo, Japan, July 3–7, 2023, Proceedings 3031362713, 9783031362712

This book constitutes the refereed proceedings of the 24th International Conference on Artificial Intelligence in Educat

1,753 79 52MB Read more

Artificial Intelligence in China: Proceedings of the 4th International Conference on Artificial Intelligence in China 9789819912568, 9789819912551, 9819912563

This book brings together papers presented at the 4th International Conference on Artificial Intelligence in China (Chin

190 42 71MB Read more