Introduction

1. Accessible Health Translation Technology Matters

2. I-D-E-A: Principles of Accessible Health Translation Technology

   1. Detectability of Machine Translation Errors to Boost User Confidence

3. Adaptability or Customizability

4. Inclusivity to Serve People of Disability

5. Equality of Accessibility Standards for Localized Multilingual Websites

1Speech and Translation Technologies: Explanations

1. 1.1Introduction

2. 1.2Automatic Speech Recognition

   1. 1.2.1Classical Automatic Speech Recognition

   2. 1.2.2Neural Automatic Speech Recognition

   3. 1.2.3Automatic Speech Recognition Issues and Directions

3. 1.3Speech Synthesis (Text-to-Speech)

   1. 1.3.1Classical Text-to-Speech

   2. 1.3.2Neural Text-to-Speech

4. 1.4Machine Translation

   1. 1.4.1Machine Translation Based on Fixed Phrases

   2. 1.4.2Full Machine Translation: Beyond Fixed Phrases

      1. 1.4.2.1Rule-Based Machine Translation

      2. 1.4.2.2Statistical Machine Translation

      3. 1.4.2.3Neural Machine Translation

5. 1.5Conclusion

2Speech and Translation Technologies: Healthcare Applications

1. 2.1Introduction

2. 2.2Obstacles to Adoption and Potential Solutions

   1. 2.2.1Reliability

      1. 2.2.1.1Offline Preparation of Output

      2. 2.2.1.2Feedback

      3. 2.2.1.3Correction

      4. 2.2.1.4Record-Keeping

   2. 2.2.2Customization per Use Case

      1. 2.2.2.1Platforms

      2. 2.2.2.2Peripherals

      3. 2.2.2.3Security

3. 2.3Speech Translation Designs for Healthcare

   1. 2.3.1Phrase-Based Speech Translation for Healthcare

      1. 2.3.1.1S-MINDS and Phraselator

      2. 2.3.1.2BabelDr

   2. 2.3.2Full Speech-to-Speech Translation for Healthcare

      1. 2.3.2.1Converser for Healthcare

      2. 2.3.2.2Fujitsu’s Focus on Ergonomics

4. 2.4Past and Current Speech Translation Systems

   1. 2.4.1Reliability of Machine Translation for Healthcare: A Study

   2. 2.4.2Surveys of Speech Translation Systems

      1. 2.4.2.1Some Bi-directional Speech Translation Systems

      2. 2.4.2.2Some Phrase-Based Speech Translation Systems

      3. 2.4.2.3Fifteen Representative Apps: A Study

      4. 2.4.2.4Some Additional Links

5. 2.5Conclusions

3Predicting Errors in Google Translations of Online Health Information

1. 3.1Introduction

2. 3.2Methods

   1. 3.2.1Research Hypothesis

   2. 3.2.2Screening Criteria for Text

   3. 3.2.3Topics of Infectious Diseases

   4. 3.2.4Labeling of Machine Translations

   5. 3.2.5Conceptual Mistakes in Machine Translations

   6. 3.2.6Prevalence of Conceptual Mistakes in Machine Translations

   7. 3.2.7Annotation of Features of English Source Texts

   8. 3.2.8Bayesian Machine Learning Classifier Relevance Vector Machine

   9. 3.2.9Training and Testing of Relevance Vector Machines with Three Different Full Feature Sets

   10. 3.2.10Classifier Optimization

   11. 3.2.11Backward Feature Elimination: RFE-SVM Method

   12. 3.2.12Separate and Joint Feature Optimization

3. 3.3Results

4. 3.4Comparison of Optimized RVMs with Binary Classifiers Using Readability Formula

5. 3.5Conclusion

4Cultural and Linguistic Bias of Neural Machine Translation Technology

1. 4.1Introduction

2. 4.2Data Collection

3. 4.3Development of Machine Learning Classifiers

4. 4.4Feature Optimization

5. 4.5Separate and Combined Feature Optimization

6. 4.6Classifier Training and Development

7. 4.7Statistical Refinement of the Optimized Classifier

8. 4.8Model Stability

9. 4.9Conclusion

5Enhancing Speech Translation in Medical Emergencies with Pictographs: BabelDr

with contributions from Pierrette Bouillon, Johanna Gerlach, Magali Norré and Herve Spechbach

1. 5.1Introduction

2. 5.2Pictographs in Medical Communication

3. 5.3BabelDr and the Bidirectional Version

4. 5.4Usability of the Bidirectional Version of BabelDr

   1. 5.4.1Patient Satisfaction

   2. 5.4.1.1Design

5. 5.4.1.2Results

   1. 5.4.2Pictograph Usability

      1. 5.4.2.1Design

6. 5.4.2.2Results

   1. 5.4.2.2.1Comprehensibility of Pictographs

   2. 5.4.2.2.2Impact of the Number and Order of Pictographic Response Choices

7. 5.5Conclusion

8. 5.6Acknowledgments

6Healthcare Accessibility for the Deaf - The BabelDr Case Study

with contributions from Irene Strasly, Pierrette Bouillon, Bastien David and Herve Spechbach

1. 6.1Introduction

2. 6.2Legal Framework in Switzerland

   1. 6.2.1Overview of the Core Principles of the Right to Health

3. 6.3Sign Language Translation Tools for Hospitals

4. 6.4BabelDr for Swiss-French Sign Language

   1. 6.4.1Recording Translations with Deaf Experts

   2. 6.4.2Virtual Avatar Generation

   3. 6.4.3Speech2sign Version of BabelDr

5. 6.5Qualitative Evaluation on the Perception of Avatars and Human Videos

6. 6.6Conclusions and Future Work

7. 6.7Acknowledgments

7Health Websites for All: A Localisation-Oriented Accessibility Evaluation

With contributions from Lucía Morado Vázquez and Silvia Rodríguez Vázquez

1. 7.1Introduction

2. 7.2Related Work

   1. 7.2.1Accessibility of Health Websites

   2. 7.2.2Localization of Health Websites

   3. 7.2.3Multilingual Web Accessibility Studies

3. 7.3Methodology

   1. 7.3.1Data Selection

   2. 7.3.2Testing Methods

   3. 7.3.3Accessibility Features Studied

      1. 7.3.3.1Language of the Page

      2. 7.3.3.2Title of the Page

4. 7.4Results

   1. 7.4.1Language of the Page

   2. 7.4.2Title of the Page

5. 7.5Discussion and Conclusions

   1. 7.5.1Challenges in Localization-Oriented Accessibility Evaluation

      1. 7.5.1.1Automated Audits

      2. 7.5.1.2Definition of Compliance Criteria

      3. 7.5.1.3Need for Accessibility Enablers with an Interdisciplinary Background

   2. 7.5.2Limitations and Future Work

6. 7.6Acknowledgments