Artificial Intelligence in Health

Name: Artificial Intelligence in Health | First International Workshop, AIH 2018, Stockholm, Sweden, July 13-14, 2018, Revised Selected Papers
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

First International Workshop, AIH 2018, Stockholm, Sweden, July 13-14, 2018, Revised Selected Papers

Fernando Koch Andrew Koster David Riaño Sara Montagna Michael Schumacher Annette ten Teije Christian Guttmann Manfred Reichert Isabelle Bichindaritz Pau Herrero Richard Lenz Beatriz López Cindy Marling Clare Martin Stefania Montani Nirmalie Wiratunga(Herausgeber*in)

Springer (Verlag)

Erschienen am 20. Februar 2019

XVI, 245 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-030-12738-1 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Additional Workshop EditorsDavid RiañoUniversitat Rovira i VirgiliTarragona, Spain Sara MontagnaUniversità di BolognaBologna, Italy Michael SchumacherUniversity of Applied Sciences and Arts Western Switzerland (HES-SO)Sierre, Switzerland Annette ten TeijeVrije Universiteit AmsterdamAmsterdam, The Netherlands Christian GuttmannNordic AI Institute, Karolinska Institute, TIETOStockholm, Sweden Manfred ReichertUlm University Ulm, Germany Richard LenzUniversity of ErlangenErlangen, Germany Beatriz Ló
Additional Workshop Editors
Preface
Organization
Acknowledgments
Contents
Agents in Health Care and Knowledge Management in Health Care
MeSHx-Notes: Web-System for Clinical Notes
1 Introduction
2 Related Work
3 MeSH Dictionary Expansion
3.1 Medical Subject Headings (MeSH)
3.2 Electronic Health Records
3.3 Word Embeddings
4 MeSHx-Notes: System Description
4.1 Back-End
4.2 Front-End
5 Term Expansion Evaluation
6 Conclusion and Further Work
References
Multiagent Systems to Support Planning and Scheduling in Home Health Care Management: A Literature Review
1 Introduction
2 Background
3 Review Methodology
3.1 Literature Search
3.2 Key Features
4 Approaches for Supporting Planning and Scheduling
5 Shortcomings of the Surveyed Approaches
6 Conclusion and Further Work
References
Ethical Surveillance: Applying Deep Learning and Contextual Awareness for the Benefit of Persons Living with Dementia
Abstract
1 Background
2 The Problem
2.1 Dementia
2.2 Privacy
2.3 A Human Rights-Based Approach
3 Machine Learning (ML)
3.1 ML Methodology
3.2 ML Results
4 Situation Appraisal
5 Contextual Factors
5.1 Time and Distance Metric
5.2 Weather Metric
6 Risk Analysis
6.1 Inferring an Unknown Location
7 Ethics
8 Conclusion
Acknowledgements
References
Active Learning for Conversational Interfaces in Healthcare Applications
1 Introduction
2 Sample Selection Methods
2.1 Query by Committee
2.2 Query by Embedded Committee
2.3 Computational Load
3 Experiments
3.1 Synthetic Example
4 Experiment with Tweets
4.1 Text Classification System
5 Results
6 Conclusions
References
Analysis of Topic Propagation in Therapy Sessions Using Partially Labeled Latent Dirichlet Allocation
1 Introduction
2 Related Work
3 Topic Modeling
4 Experimental Dataset
4.1 Data Description
4.2 Preprocessing of the Meta-data
4.3 Preprocessing of the Conversation Text
5 The Proposed Approach for TDT
5.1 Topic Modeling
5.2 Topic Inference
5.3 Tracking Topics
6 Evaluation
7 Conclusions
References
Dr. AI, Where Did You Get Your Degree?
1 Introduction
2 Regulatory Design for AIaMD
2.1 Accrediting Our Data Sources and Methods
2.2 Focus on the Outcomes
3 So Can We Treat AI Like a Doctor?
4 Failure Points
4.1 Recalling AI
4.2 Adversarial AI and Security
5 Conclusion
References
Design Principles and Action Reflection for Agent-Based Assistive Technology
1 Introduction
2 Theoretical Background
2.1 Activity Theory
2.2 Formal Argumentation Theory
3 Reflection on Decisions About Human Activity
3.1 Support in Relation to the Zone of Proximal Development Using Formal Argumentation
4 Principles for Providing Consistent Assistive Services
4.1 Activity-Oriented Principles
5 Implementation
6 Discussion and Conclusions
References
Microsoft Hololens - A mHealth Solution for Medication Adherence
1 Introduction
2 Medication Scenario
2.1 Patient Groups
2.2 Rules for Interchangeable Medicines
2.3 Prescription and On Demand Medicines
3 Research Questions and Methodology
3.1 Maintaining Self Management
3.2 Confirmation and Patient Security
3.3 Personalization
3.4 Mobility
3.5 Replacement for Nurses
4 Theoretical Framework
4.1 Data Modeling
4.2 Decision Making Modeling
5 Implementation
6 Evaluation
7 Related Work
8 Conclusions and Future Work
References
A Knowledge-Based Simulation Framework for Decision Support in Brazilian National Cancer Institute
Abstract
1 Introduction
2 Methods
3 INCA Knowledge-Based Simulation Framework
3.1 Clinical Analysis (Patient Treatment Flow)
3.2 Capacity Analysis (Simulation Model)
4 Results
5 Conclusions
References
Data Science and Decision Systems in Medicine
Lifted Maximum Expected Utility
1 Introduction
2 Related Work
3 Parameterised Probabilistic Models
4 Lifted Maximum Expected Utility
4.1 Parameterised Probabilistic Decision Models
4.2 Maximum Expected Utility
4.3 How to Model Utilities in a Medical Context
5 Solving the MEU Problem and Answer Multiple Marginal Queries Efficiently
5.1 Lifted Junction Tree Algorithm
5.2 meuLJT
6 Conclusion
References
The Role of Usability Engineering in the Development of an Intelligent Decision Support System
1 Introduction
1.1 Healthcare Technology Acceptance
1.2 The PEPPER System
1.3 User-Centred Design
1.4 Paper Scope
2 Methods
2.1 User Research
2.2 Analysis
2.3 Iterative Design and Formative Evaluation
2.4 Summative Evaluation
3 Results
3.1 User Research
3.2 Analysis
3.3 Iterative Design and Formative Evaluation
3.4 Summative Evaluation
4 Discussion
5 Conclusion
References
Automated Pain Detection in Facial Videos of Children Using Human-Assisted Transfer Learning
1 Introduction
2 Methods
2.1 Participants
2.2 Experimental Design and Data Collection
2.3 Feature Extraction
2.4 Machine Learning Models
3 Analysis and Discussion
3.1 Automated Classifier Performance Varies by Environment
3.2 Classification Based on Manual AUs Are Less Sensitive to Environmental Changes
3.3 Restricting Manual AUs to Those Associated with Pain Improves Classification
3.4 iMotions AUs Are Different Than Manual FACS AUs
3.5 Transfer Learning via Mapping to Manual Features Improves Performance
4 Results
4.1 Test on New Subjects with only iMotions AU Codings
4.2 Test with Masked Pain and Faked Pain
5 Conclusion
6 Future Work
References
Towards Automated Pain Detection in Children Using Facial and Electrodermal Activity
1 Introduction
2 Methods
2.1 Participants
2.2 Experimental Design and Data Collection
2.3 Feature Extraction and Processing
2.4 Machine Learning Models
2.5 Evaluation Metrics
3 Results and Discussion
3.1 Performance Using Video/EDA Features
3.2 Fusion of Video and EDA
3.3 Training with V2 Scores
3.4 Transfer Learning for Video Features
3.5 Fusion of Transferred Video and EDA
4 Conclusion
References
Interpretation of Best Medical Coding Practices by Case-Based Reasoning-A User Assistance Prototype for Data Collection for Cancer Registries
1 Introduction
2 Case-Based Interpretation of Best Practices
2.1 Preliminaries
2.2 Global Architecture
2.3 Retrieve
2.4 Reuse
2.5 Revise and Retain
3 Prototype and Preliminary Results
4 Conclusion
References
Identification of Serious Illness Conversations in Unstructured Clinical Notes Using Deep Neural Networks
1 Introduction and Related Work
2 Data
2.1 Data Source
2.2 Cohort
2.3 Clinical Domains
2.4 Annotation
3 Methods
3.1 Pre-processing
3.2 Regular Expression
3.3 Artificial Neural Network
4 Results
4.1 Evaluation Metrics
4.2 Performance
4.3 Error Analysis
4.4 Effect of Training Set Size
5 Discussion and Future Work
6 Conclusion
A Regular Expression Library
B Token-Level Performance
C Examples of Identified Text
References
Generating Reward Functions Using IRL Towards Individualized Cancer Screening
1 Introduction
2 Background
3 Materials and Methods
3.1 NLST Dataset
3.2 Athena Dataset
3.3 Partially Observable Markov Decision Processes
3.4 Maximum Entropy IRL
3.5 Adaptive Step Size
3.6 Computation of Rewards
4 Evaluation and Results
4.1 Comparison of MaxEnt IRL with and Without Adaptive Step Size
4.2 Lung and Breast POMDP Results
5 Discussion
References
Deep Learning Architectures for Vector Representations of Patients and Exploring Predictors of 30-Day Hospital Readmissions in Patients with Multiple Chronic Conditions
Abstract
1 Introduction
2 Related Work
3 Methods
3.1 Study Design
3.2 Data Collection
4 Results
4.1 Experiment 1: CNN for Vector Representations of MCC Patients with 30-Day Hospital Readmissions
4.2 Experiment 2: RETAIN for MCC Patients with 30-Day Hospital Readmissions
5 Discussion
5.1 Effectiveness of CNN and RNN for MCC Patients
5.2 Sub-typing and Vector Representations of MCC Patients
5.3 Practical Implications for MCC Patients and Health Care in General
6 Limitations/Methodological Considerations
7 Conclusion
Funding
Ethical Considerations
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken