AI for Disease Surveillance and Pandemic Intelligence: Intelligent Disease Detection in Action (Studies in Computational Intelligence, 1013)

Preface
Contents
Contributors
Abbreviations
Digital Technologies for Clinical, Public and Global Health Surveillance
	1 Introduction
	2 Artificial Intelligence-Based Tools and Methods
	References
Imputing Fine-Grain Patterns of Mental Health with Statistical Modelling of Online Data
	1 Introduction
	2 Methods
		2.1 Regression Analysis
	3 Results
		3.1 Web Search Queries as Predictors of Annual Suicide Rates
		3.2 Imputing Monthly Near-Real-Time Resolution from Annual Ground Truth Data
		3.3 Prediction
	4 Discussion
	5 Conclusion
	References
Lexical and Acoustic Correlates  of Clinical Speech Disturbance  in Schizophrenia
	1 Introduction
	2 Related Work
	3 Methods
		3.1 Dataset
		3.2 Data Processing and Analysis
	4 Results
	5 Discussion
		5.1 Limitations and Future Work
	6 Acknowledgements and Conflicts of Interest
	References
A Prognostic Tool to Identify Youth at Risk of at Least Weekly Cannabis Use
	1 Introduction
	2 Methods
	3 Results
	4 Discussion
	References
Neuro-Symbolic Neurodegenerative Disease Modeling as Probabilistic Programmed Deep Kernels
	1 Introduction
	2 Background
		2.1 Probabilistic Programming
		2.2 Neurodegenerative Disease Modeling
	3 Problem Formulation
	4 Proposed Method
		4.1 Monotonic Gaussian Processes
		4.2 Probabilistic Programmed Deep Kernel Learning
		4.3 Neurodegeneration Programs
	5 Experiments
		5.1 Results
	6 Conclusion
	References
Self-Disclosure in Opioid Use Recovery Forums
	1 Introduction
	2 Related Work
	3 Dataset
	4 Self-disclosure Expressed in Post-titles
		4.1 Features
		4.2 Predicting Self-Disclosure in Opioid Use Recovery Forums
	5 Social Support Expressed in Post-titles
	6 What Type of Social Support Is Sought in Post-titles that Disclose Positive or Negative Information
		6.1 Results
	7 Accountability
	8 Discussion
	9 Conclusion and Future Work
	References
Identifying Prepubertal Children with Risk for Suicide Using Deep Neural Network Trained on Multimodal Brain Imaging
	1 Introduction
	2 Previous Work
	3 Dataset and Preprocessing
		3.1 Adolescent Brain and Cognitive Development Study
	4 Methods
		4.1 Deep Neural Network
		4.2 Model Training and Testing
		4.3 Experiment Settings
	5 Results
	6 Discussion
	References
Improving Adverse Drug Event Extraction with SpanBERT on Different Text Typologies
	1 Introduction
	2 Related Work
	3 The Proposed Approach
	4 Experimental Results
		4.1 Datasets
		4.2 Metrics
		4.3 Baselines
		4.4 Implementation Details
		4.5 Evaluation
		4.6 Quantitative Analysis
		4.7 Qualitative Error Analysis
	5 Conclusions
	References
Machine Learning Identification of Self-reported COVID-19 Symptoms from Tweets in Canada
	1 Introduction
	2 Methodology
	3 Results and Discussion
	4 Conclusion
	References
RRISK: Analyzing COVID-19 Risk in Food Establishments
	1 Background and Motivation
	2 Related Research
	3 Application Overview
	4 Approach
	5 Data
	6 Algorithm
		6.1 Data Statistics
	7 System Architecture
	8 Training, Validation, and Verification
		8.1 Similar Restaurants with Uncorrelated Yelp and RRISK Scores
		8.2 A Risky Region with a Reasonable Restaurant
	9 Implementation Challenges
	10 Conclusions and Future Work
	References
AWS CORD-19 Search: A Neural Search Engine for COVID-19 Literature
	1 Introduction
	2 System Overview
		2.1 Amazon Kendra
		2.2 Comprehend Medical
		2.3 COVID-19 Knowledge Graph
		2.4 Topic Models
	3 Evaluation
		3.1 Paper Recommendation
	4 Analysis
	5 Limitations and Future Directions
	6 Conclusion
	References
Inferring COVID-19 Biological Pathways from Clinical Phenotypes Via Topological Analysis
	1 Introduction
	2 Background
		2.1 Redescriptions
		2.2 Topological Data Analysis
	3 Proposed Pipeline
	4 Experimental Setup
		4.1 Dataset
		4.2 Experimental Details
	5 Results
		5.1 Main Result
		5.2 Discussion
	6 Conclusions
	References
The EpiBench Platform to Propel AI/ML-Based Epidemic Forecasting: A Prototype Demonstration Reaching Human Expert-Level Performance
	1 Introduction
	2 Background
		2.1 Existing Infrastructure
		2.2 Acceptable Methods
	3 Results from the Prototype
		3.1 Ensemble Development
		3.2 Assessing Forecasting Decisions
	4 Discussions
		4.1 Evaluation Protocol
		4.2 EpiBench: Planned Functionalities
	5 Conclusions
	References
Interpretable Classification of Human Exercise Videos Through Pose Estimation and Multivariate Time Series Analysis
	1 Introduction
	2 Related Work
	3 Data Collection
	4 Methods
	5 Experiments
		5.1 Classifier Accuracy
		5.2 Classifier Runtime
		5.3 Classifier Feedback
	6 Conclusion
	References
Interpreting Deep Neural Networks for Medical Imaging Using Concept Graphs
	1 Introduction
	2 Proposed Framework
		2.1 Concept Formation
		2.2 Concept Identification
		2.3 Network Formation and Information Flow
		2.4 Trail Estimation
	3 Experiments
		3.1 Brain Tumor Segmentation
		3.2 Diabetic Retinopathy Classification
	4 Related Work
	5 Discussion
	6 Appendix 1
	References
Do Deep Neural Networks Forget Facial Action Units?—Exploring the Effects of Transfer Learning in Health Related Facial Expression Recognition
	1 Introduction
	2 Related Work
		2.1 Transfer Learning in Pain Recognition
		2.2 Explainable Artificial Intelligence
	3 Pain Training
		3.1 Datasets
		3.2 Transfer Learning
	4 Measuring Forgetting
		4.1 Re-Train for Measuring Forgetting
		4.2 Visual Analysis
		4.3 Concept Embedding Detection
	5 Results
	6 Discussion
	7 Conclusion
	References
Utilizing Predictive Analysis to Aid Emergency Medical Services
	1 Introduction
	2 Data Source and Preparation
	3 Model Development and Evaluation
	4 Results
	5 Discussion
		5.1 Usability of the System
		5.2 Scope of the System
		5.3 Limitations
	6 Conclusion
	7 Future Work
	References
Measuring Physiological Markers of Stress During Conversational Agent Interactions
	1 Introduction
	2 Methods
		2.1 Study Designs
		2.2 Participants
		2.3 Procedures
	3 Analysis
		3.1 Data Preprocessing
		3.2 Statistical Analyses
	4 Results
		4.1 External Validation of the CA Study Resting Period
		4.2 Signal Comparisons Between Tasks
		4.3 Comparing the Rest Period and CA Interaction
		4.4 Comparing the CA and WESAD Studies
		4.5 HRV Analysis
	5 Discussion
	References
EvSys: A Relational Dynamic System for Sparse Irregular Clinical Events
	1 Introduction
	2 EvSys for Clinical Events
		2.1 Preliminaries
		2.2 Disentangling Measurements from Health States
		2.3 Dynamic of Measurement Processes
		2.4 Dynamic of Health Processes with Skipping
		2.5 Predictions
	3 Experiments
		3.1 Evaluation Setup
		3.2 Results
	4 Related Works
	5 Conclusion
	References
Predicting Patient Outcomes with Graph Representation Learning
	1 Introduction
	2 Related Work
	3 Methods
	4 Data
	5 Results
	6 Discussion
	References
Patient-Specific Seizure Prediction Using Single Seizure Electroencephalography Recording
	1 Introduction
	2 Related Work
	3 Motivation
	4 Methodology
		4.1 Data and Preprocessing
		4.2 Seizure Prediction Using Siamese Learning
	5 Evaluation
	6 Conclusion and Future Work
	References
Evaluation Metrics for Deep Learning Imputation Models
	1 Introduction
	2 Imputation Models and Evaluation Metrics
		2.1 Imputation Models
		2.2 Evaluation Metrics
		2.3 Evaluation Methodology
	3 Comparative Analysis
		3.1 Data Collection and Processing
		3.2 Evaluation Procedure
		3.3 Results
	4 Limitations of Evaluation Metrics
	5 Related Work
	6 Conclusion and Future Work
	References
Logistic Regression is also a Black Box. Machine Learning Can Help
	1 Introduction
	2 Materials and Methods
		2.1 Data Simulation Protocol
		2.2 Classification
		2.3 Classification Performance Evaluation
	3 Results
	4 Discussion
	5 Conclusions
	References