Luxton Artificial Intelligence in Behavioral and Mental Health Care

1. Auflage 2015
ISBN: 978-0-12-800792-1
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, 308 Seiten

Artificial Intelligence in Behavioral and Mental Health Care summarizes recent advances in artificial intelligence as it applies to mental health clinical practice. Each chapter provides a technical description of the advance, review of application in clinical practice, and empirical data on clinical efficacy. In addition, each chapter includes a discussion of practical issues in clinical settings, ethical considerations, and limitations of use. The book encompasses AI based advances in decision-making, in assessment and treatment, in providing education to clients, robot assisted task completion, and the use of AI for research and data gathering. This book will be of use to mental health practitioners interested in learning about, or incorporating AI advances into their practice and for researchers interested in a comprehensive review of these advances in one source. - Summarizes AI advances for use in mental health practice - Includes advances in AI based decision-making and consultation - Describes AI applications for assessment and treatment - Details AI advances in robots for clinical settings - Provides empirical data on clinical efficacy - Explores practical issues of use in clinical settings

Autoren/Hrsg.

Luxton, David D.

Inhaltsverzeichnis

1;Front Cover;1
2;Artificial Intelligence in Behavioral and Mental Health Care;4
3;Copyright Page;5
4;Contents;6
5;List of Contributors;10
6;About the Editor;12
7;Preface;14
8;1 An Introduction to Artificial Intelligence in Behavioral and Mental Health Care;16
8.1;Introduction and Overview;16
8.2;Key Concepts and Technologies;17
8.2.1;What Is AI?;17
8.2.2;Machine Learning and Artificial Neural Networks;18
8.2.3;Natural Language Processing;20
8.2.4;Machine Perception and Sensing;22
8.2.5;Affective Computing;22
8.2.6;Virtual and Augmented Reality;23
8.2.7;Cloud Computing and Wireless Technologies;24
8.2.8;Robotics;24
8.2.9;BCIs and Implants;25
8.2.10;Supercomputing and Brain Simulation;26
8.2.11;The Turing Test;28
8.2.12;Technological Barriers;29
8.3;Benefits of AI for Behavioral and Mental Health Care;30
8.3.1;Intelligent Machines Are Better at Some Things;30
8.3.2;Improved Self-Care and Access to Care;30
8.3.3;Integration and Customization of Care;31
8.3.4;Economic Benefits;31
8.4;Additional Considerations;32
8.5;Conclusion;34
8.6;References;36
8.7;Additional Resources;40
9;2 Expert Systems in Mental Health Care: AI Applications in Decision-Making and Consultation;42
9.1;Introduction;42
9.2;The History – Expert Systems and Clinical Artificial Intelligence in Health Care;43
9.3;The Present – Dynamical Approaches to Clinical AI and Expert Systems;47
9.3.1;Temporal Modeling Overview;47
9.3.2;Real-World Clinical Applications – Predicting in a Dynamic World;47
9.3.3;Multi-Agent Models for Personalized Medicine;51
9.4;Technology-Enhanced Clinicians;52
9.5;Summary of Dynamical Approaches for Clinical AI;53
9.6;The Future;54
9.6.1;Cognitive Computing in Health Care;54
9.6.2;The Intersection Between Other Emerging Technologies and Clinical Artificial Intelligence;56
9.6.3;Ethics and Challenges;60
9.7;Conclusion;61
9.8;References;62
10;3 Autonomous Virtual Human Agents for Healthcare Information Support and Clinical Interviewing;68
10.1;Introduction;68
10.2;The Rationale and Brief History of the Clinical Use of VHs;70
10.3;Use Cases: SimCoach and SimSensei;74
10.3.1;SimCoach: A VH Agent to Support Healthcare Information Access;74
10.3.2;SimSensei: A VH Interviewing Agent for Detection and Computational Analysis of Psychological Signals;79
10.3.2.1;Nonverbal Behavior and Clinical Conditions;82
10.4;Comparative Evaluation Across Interviews: Face-To-Face, WoZ, and Automatic Interaction with the SimSensei VH Agent;85
10.5;Conclusions;89
10.6;References;90
11;4 Virtual Affective Agents and Therapeutic Games;96
11.1;Introduction;96
11.2;Brief History of Virtual Affective Agents and Serious Games;99
11.2.1;Virtual Affective Agents;99
11.2.2;Serious Games;102
11.3;State of the Art;103
11.3.1;Nonverbal Interaction: Emotion Recognition, Emotion Expression, and Agent Embodiment;104
11.3.1.1;Emotion Recognition and Expression;104
11.3.1.2;Embodiment;106
11.3.2;Believability, Affective Realism, and Emotional Intelligence;107
11.3.3;Personalization and Adaptation Capabilities;108
11.3.4;Putting It All Together: Agent Architectures;109
11.3.5;Nonplaying Characters and Player Avatars;112
11.3.6;Affective and Affect-Adaptive Gaming;113
11.3.7;Overview of Recent Applications in Health Care;113
11.3.8;Pediatric Pain Management – “Free Dive”;115
11.3.9;OCD in Children – “Ricky and the Spider”;116
11.3.10;PTSD in War Veterans – “Virtual Iraq”;116
11.3.11;Social and Emotion Regulation Skills for Children on the Autism Spectrum – “Secret Agent Society”;117
11.4;Applicable Ethical and Privacy Considerations;119
11.4.1;Affective Privacy;119
11.4.2;Emotion Induction;120
11.4.3;Virtual Relationships;121
11.5;Future Prospects;122
11.5.1;Proliferation;123
11.5.2;Formal Evaluations;123
11.5.3;Improved Understanding of Suitable Applications and Contexts for Agents and Games;123
11.5.4;Agents: Empathy and Personality;124
11.5.5;Improved User State Recognition, Affective User Modeling and Personalization;124
11.5.6;Natural Language Understanding;125
11.5.7;New Types of Relationships and Improved Understanding of Relationships;125
11.6;Conclusions;126
11.7;References;126
12;5 Automated Mental State Detection for Mental Health Care;132
12.1;Introduction;132
12.2;Theoretical and Technical Foundation;134
12.3;Example Systems;137
12.3.1;Affective States;138
12.3.1.1;Basic Emotions;138
12.3.1.2;Nonbasic Emotions;139
12.3.1.3;Affect Dimensions;140
12.3.2;Attentional Lapses (Mind Wandering);141
12.3.3;Pain;143
12.3.4;Depression;143
12.3.5;Stress;144
12.4;Concluding Remarks;146
12.5;Acknowledgments;147
12.6;References;147
13;6 Intelligent Mobile, Wearable, and Ambient Technologies for Behavioral Health Care;152
13.1;Introduction;152
13.1.1;Overview of Intelligent Mobile Health;153
13.2;Intelligent Capabilities for Mobile Health;154
13.2.1;Mobile Sensors;155
13.2.2;Example Applications of Multi-Modal Sensor Technology in Smart Mobile Health;157
13.2.3;Speech Recognition and NLP;158
13.2.4;Examples of Speech Recognition Technologies in Health Care;159
13.2.5;Virtual Humans on Mobile Platforms;160
13.2.6;Examples of Virtual Human Health Interventions on Mobile Platforms;162
13.2.7;Augmented Reality on Mobile Devices;163
13.2.8;Example AR Applications in Behavioral Health Care;163
13.2.9;Summary;164
13.3;Overview of AmI;165
13.3.1;Example of AmI for Behavioral Health Applications;165
13.3.2;The Internet of Things;167
13.4;Design Recommendations;167
13.4.1;Mobile Health Design Considerations;168
13.4.2;AmI Design Considerations;169
13.4.3;Privacy, Data Security, and Ethics Considerations;170
13.5;Conclusion;171
13.6;References;172
14;7 Artificial Intelligence and Human Behavior Modeling and Simulation for Mental Health Conditions;178
14.1;Introduction;178
14.2;Background;178
14.2.1;Why ABMS and Challenges;181
14.3;History of ABMS in Medicine/Mental Health Care;181
14.4;Synergies with Other Industries;182
14.5;Sociological Inputs into Multi-Tiered ABMS;182
14.5.1;Antonovksy’s Salutogenesis Model;182
14.5.2;Theory of Reasoned Action;183
14.5.3;Social Factor Impact on Readmission or Mortality;183
14.6;A Toolbox for Multi-Layer Modeling of Social Systems;184
14.6.1;Agent Mind–Body Level: The PMFserv Architecture;184
14.6.1.1;Agent Motives;185
14.6.1.2;Agent State Properties;185
14.6.2;Organizational Level: StateSim;187
14.6.3;Societal Level: StateSim;189
14.7;Data and Privacy Constraints for ABMS in Mental Health Modeling Applications;190
14.7.1;National Data Sources;191
14.7.2;National Survey on Drug Use and Health (NSDUH);191
14.7.3;Local Data Sources;191
14.7.4;Penn Data Warehouse;191
14.7.5;US Census and GIS Data;192
14.8;Example Application;192
14.8.1;Agent/Individual Modeling;193
14.8.2;Organization Modeling;193
14.8.3;Population Modeling;194
14.9;Future Prospects;195
14.10;Conclusion;196
14.11;Acknowledgments;196
14.12;References;197
15;8 Robotics Technology in Mental Health Care;200
15.1;Introduction;200
15.1.1;Background;201
15.1.1.1;Human–Robot Interaction;201
15.1.1.2;Robot Morphology;201
15.1.1.3;Robot Capabilities;203
15.1.1.4;Robot Autonomy;203
15.1.2;Recent Applications;204
15.1.2.1;Autism Spectrum Disorders;204
15.1.2.2;Activity Engagement and Physical Exercise;205
15.1.2.3;Dementia and Age-Related Cognitive Decline;206
15.1.2.4;Companion Robots to Improve Psychosocial Outcomes;207
15.1.2.5;Clinician Training for Interacting with People with Disabilities;207
15.1.2.6;Diagnosing and Studying Schizophrenia;209
15.1.3;Design Issues;209
15.1.3.1;Potential Barriers to Provider Technology Adoption;209
15.1.3.2;Cultural Barriers to Technology Adoption;210
15.1.3.3;A Need for Evidence-Based Robotics Use in Mental Health Care;211
15.1.4;Ethical Issues;212
15.1.4.1;The Prime Directive;212
15.1.4.2;Specific Principles;213
15.1.4.2.1;Human Dignity Considerations;213
15.1.4.2.2;Design Considerations;213
15.1.4.2.3;Legal Considerations;213
15.1.4.2.4;Social Considerations;213
15.2;Conclusion;214
15.3;Acknowledgment;215
15.4;References;215
16;9 Public Health Surveillance: Predictive Analytics and Big Data;220
16.1;Introduction;220
16.2;The Current State of Informatics;221
16.3;Overview of Recent Applications;221
16.3.1;Data Workflow;222
16.3.2;The Durkheim Project;224
16.3.3;Background;225
16.3.4;Related Work;226
16.3.5;Overview;226
16.4;Results;230
16.4.1;Implications;230
16.5;Larger Cohorts (Current Work);233
16.5.1;The Durkheim Project Architecture;233
16.6;Impact;235
16.7;Applicable Ethical Considerations;237
16.7.1;Consent;238
16.7.2;Privacy;238
16.7.3;Transparency;238
16.7.4;Discussion;239
16.8;Future Prospects in the Topic Area;239
16.8.1;Next-Generation Inference;239
16.8.2;Approaches in Deep Learning;241
16.8.3;Limitations of Deep Learning;241
16.8.4;Latest Extensions of Deep Learning;241
16.8.5;Technical Steps, Challenges, and Risks;242
16.8.6;Deep Learning, a Discussion;242
16.9;Conclusion;243
16.10;References;243
17;10 Artificial Intelligence in Public Health Surveillance and Research;246
17.1;Introduction;246
17.1.1;Living in a Dark Tunnel of Pain: Automatic Screening for Depression;248
17.1.2;Neurodegenerative Diseases and the e-Health Challenge;254
17.1.3;Mass Shooters and the Challenge of Screening for Harmful Offenders;257
17.1.4;Vectorial Semantics and Personality Analysis;259
17.1.5;Analyzing the Texts of the Mass Shooters;261
17.1.6;Preprocessing of the Text and Analysis;262
17.1.7;Analysis and Results;264
17.1.8;The Screening Procedure;264
17.2;Conclusion;266
17.3;References;267
18;11 Ethical Issues and Artificial Intelligence Technologies in Behavioral and Mental Health Care;270
18.1;Introduction;270
18.2;Overview of Ethics Codes and Ethical Behavior in Health Care;271
18.2.1;Background;271
18.2.2;Technology-Associated Ethics Codes and Guidelines;273
18.3;Particular Ethics Challenges;276
18.3.1;Therapeutic Relationships and Emotional Reactions;276
18.3.2;Competence of Intelligent Machines;278
18.3.3;Patient Safety;280
18.3.4;Respect of Privacy and Trust;281
18.3.5;Deception and Appearance;283
18.3.6;Responsibility;283
18.4;Design and Testing Recommendations;285
18.4.1;Ethical (Moral) Turing Test;285
18.4.2;GenEth: A General Ethical Dilemma Analyzer;286
18.4.3;What Can Ethical Machines Teach Us?;288
18.5;Conclusion;289
18.6;References;290
19;Glossary;292
20;Index;296
21;Back Cover;309

Leseproben

Chapter 2

Expert Systems in Mental Health Care

AI Applications in Decision-Making and Consultation

Casey C. Bennett1,2 and Thomas W. Doub2,    1School of Informatics and Computing, Indiana University, Bloomington, IN, USA,    2Department of Informatics, Centerstone Research Institute, Nashville, TN, USA

Artificial intelligence (AI) based tools hold potential to extend the current capabilities of clinicians, to deal with complex problems and ever-expanding information streams that stretch the limits of human ability. In contrast to previous generations of AI and expert systems, these approaches are increasingly dynamical and less computationalist – less about “rules” and more about leveraging the dynamic interplay of action and observation over time. The (treatment) choices we make change what we observe (clinically, or otherwise), which changes future choices, which affects future observations, and so forth. As humans (clinicians or otherwise), we leverage this fact every day to act “intelligently” in our environment. To best assist us, our clinical computing tools should approximate the same process. Such an approach ties to future developments across the broader healthcare space, e.g., cognitive computing, smart homes, and robotics.

Keywords

Artificial intelligence; medical decision-making; expert systems; mental health; health care; clinical decision support systems; cognition; cognitive computing; temporal modeling; dynamical systems

Introduction

Across real-world scenarios – clinical ones included – perceptions (e.g., observations) and actions (e.g., treatments) are structured in a circular fashion (Merleau-Ponty, 1945). The actions we take change what we perceive in the future, and in turn those perceptions may alter the future actions we take (isomorphic to changes in the human visual system due to movement in the world (Gibson, 1979)). There is inherent in this dynamical process. As humans we leverage this fact every day to act “intelligently” in our environment. We think about problems in a temporally extended fashion, whether it be during treatment of a patient or making a left turn in our car. For instance, when driving a car we don’t simply decide to make a left turn and then do it. Rather, there is constant perceptual feedback (e.g., if a pedestrian suddenly appears in the crosswalk, we adjust our actions). This alters further perceptions; we may alter our turning radius to avoid said pedestrian, which results in finding a fire hydrant directly in our path. Given the probability of such a sequence (e.g., how busy the pedestrian traffic is at the crosswalk), we may choose not to turn at the intersection at all, or find an alternate route. The point is that our actions lead to certain perceptions that we use to make decisions. The same is true for clinical decision-making. We are not merely passive observers of “data” – data is a process of interaction. Should not our clinical computing tools approximate the same process? If we want tools to enhance our cognition and/or improve our decision-making, those tools need to fit the way we think about the world. In other words, they should provide a sort of that enables people to do what they do better (Clark, 2004, 2013; Sterelny, 2007).

In this chapter, we describe emerging approaches for doing exactly that, i.e., temporal modeling. Such approaches are ripe for application to health care, where treatment decisions must be made over time, and where continually reevaluating ongoing treatment is critical to optimizing clinical care for individual patients. This is especially true for chronic health conditions, such as mental illness, which forms the bulk of healthcare expenditures in the United States (Orszag & Ellis, 2007). Clinicians do not just make decisions and move forward – rather they are constantly reevaluating those decisions, titrating medications, adjusting treatments, making new observations. It is a very dynamic process, both in terms of the treatment being delivered as well as the cognitive processes of the clinician and patient (e.g., how they integrate information into their decision-making over time (Patel, Kaufman, & Arocha, 2002)). This represents a major obstacle, because currently much of the focus of AI and clinical decision support systems (CDSS) in healthcare is on making a single recommendation at a single timepoint. But that is not how health care really works.

This chapter is laid out as follows. “The History – Expert Systems and Clinical Artificial Intelligence in Health Care” provides a brief history of AI, expert systems, and CDSS in physical and mental healthcare. The successes and failures of such efforts lead into “The Present – Dynamical Approaches to Clinical AI and Expert Systems,” where we discuss current research around artificial intelligence in healthcare, including dynamical approaches that explicitly incorporate time. In “The Future,” we expand upon this current work to detail future directions around how such approaches can integrate into the broader healthcare space, for example, cognitive computing, smart homes, cyborg clinicians, and robotics. We conclude with a discussion of what this all may mean for the future of health care, mental health, and clinicians and patients alike. AI is a term often loosely applied, with “intelligence” being more of a romantic notion than a precise descriptor (Brooks, 1991). But all is not lost. The aim of this chapter is to help readers understand where we have been, where we are, and where we are going in our ongoing quest to put “intelligence” into AI, for clinical applications and beyond.

The History – Expert Systems and Clinical Artificial Intelligence in Health Care

Efforts to develop AI, both within and outside of health care, have a long history. Some of the earliest successful applications of AI in health care were (Jackson, 1998; Luxton, 2014). An expert system is a computer system that is designed to emulate the decision-making capabilities and performance of human experts. Traditionally, this was done by eliciting a knowledge base of rules from experts (), from which inference about the present state or future could be performed () by an end-user (via a ), as shown in Figure 2.1. The rules often took the form of “if-then”, where the “then” component typically comprised a probability. For instance, the patient has symptom , the probability of disease is, say, 0.6. A multitude of such rules could then be used to calculate probabilistic recommendations.

Figure 2.1 Basic outline of an expert system.

One well-known early example of an expert system in health care was MYCIN, developed in the 1970s at Stanford University. The system was designed to identify bacterial infections and recommend appropriate antibiotic treatment (Shortliffe, 1976). Similar developments were also underway at the same time on the mental health side. For instance, DIAGNO was an early tool for computer-assisted psychiatric diagnosis that was developed at Columbia University in the 1960s and 1970s. It used as input 39 clinical-observation scores processed through a decision tree, resulting in a differential diagnosis. The system achieved comparable performance as human clinicians across a variety of mental disorders (Spitzer & Endicott, 1974), though it was never put to use in real-world practice.

Subsequent years saw the inclusion of expert systems into many CDSS. Decision support, as the name implies, refers to providing information to clinicians, typically at the point of decision-making (Osheroff et al., 2007). However, we should be careful to point out that not all CDSS tools are necessarily expert systems or AI – many are simply hard-coded rules that trigger alerts or messages, containing neither probabilistic rules nor inferential reasoning. Nonetheless, some CDSS tools do embody principles of expert systems. One recent example in mental health care is the TMAP project from UT-Southwestern Medical School for computer-assisted depression medication treatment (Shelton & Trivedi, 2011; Trivedi et al., 2009; Trivedi, Kern, Grannemann, Altshuler, & Sunderajan, 2004). The system used algorithms to suggest appropriate changes to medications and/or dosing via electronic health record systems. It worked well in research studies, though it faced various implementation challenges in practice (see “Ethics and Challenges” section below).

CDSS tools – both those based on expert system models and otherwise – have had a mixed history of success (Garg et al., 2005; Jaspers, Smeulers, Vermeulen, & Peute, 2011; Kawamoto, Houlihan, Balas, & Lobach, 2005). Many are based on evidence-based guidelines (typically derived from expert opinion or statistical averages) that prescribe a one-size-fits-all treatment regimen for every patient, or a standardized sequence of treatment options (Bauer, 2002; Bennett, Doub, & Selove, 2012; Green, 2008). However, real-world patients display individualized characteristics and symptoms that impact treatment effectiveness. As such, clinicians quickly learn to ignore recommendations that say...