This book takes an in-depth look at the emerging technologies that are transforming the way clinicians manage patients, while at the same time emphasizing that the best practitioners use both artificial and human intelligence to make decisions.
AI and machine learning are explored at length, with plain clinical English explanations of convolutional neural networks, back propagation, and digital image analysis. Real-world examples of how these tools are being employed are also discussed, including their value in diagnosing diabetic retinopathy, melanoma, breast cancer, cancer metastasis, and colorectal cancer, as well as in managing severe sepsis.
With all the enthusiasm about AI and machine learning, it was also necessary to outline some of criticisms, obstacles, and limitations of these new tools. Among the criticisms discussed: the relative lack of hard scientific evidence supporting some of the latest algorithms and the so-called black box problem. A chapter on data analytics takes a deep dive into new ways to conduct subgroup analysis and how it’s forcing healthcare executives to rethink the way they apply the results of large clinical trials to everyday medical practice. This re-evaluation is slowly affecting the way diabetes, heart disease, hypertension, and cancer are treated. The research discussed also suggests that data analytics will impact emergency medicine, medication management, and healthcare costs.
An examination of the diagnostic reasoning process itself looks at how diagnostic errors are measured, what technological and cognitive errors are to blame, and what solutions are most likely to improve the process. It explores Type 1 and Type 2 reasoning methods; cognitive mistakes like availability bias, affective bias, and anchoring; and potential solutions such as the Human Diagnosis Project. Finally, the book explores the role of systems biology and precision medicine in clinical decision support and provides several case studies of how next generation AI is transforming patient care.
The book explains to physicians and technologists the value and limitations of artificial intelligence in the management of disease. Specifically, it explains how machine learning and new types of data analysis will improve diagnosis and personalize patient care.
| Dedication |
|
vii | |
| Contents |
|
ix | |
| Preface |
|
xiii | |
| Reinventing CDS Requires Humility in the Face of Overwhelming Complexity |
|
xiii | |
| References |
|
xvi | |
| About the Authors |
|
xvii | |
|
Chapter 1 Clinical Reasoning and Diagnostic Errors |
|
|
1 | (24) |
|
Measuring Diagnostic Errors |
|
|
1 | (4) |
|
Understanding the Multiple Causes of Diagnostic Errors |
|
|
5 | (1) |
|
|
|
6 | (1) |
|
Type 1 and Type 2 Thinking |
|
|
7 | (1) |
|
Combining Cognitive Approaches |
|
|
8 | (3) |
|
|
|
11 | (2) |
|
Improving Clinicians' Diagnostic Skills |
|
|
13 | (4) |
|
Listening More, Talking Less |
|
|
17 | (3) |
|
In Clinical Decision Support, Practical Beats Sexy |
|
|
20 | (1) |
|
|
|
21 | (4) |
|
Chapter 2 The Promise of Artificial Intelligence and Machine Learning |
|
|
25 | (34) |
|
Defining Terms, Understanding Concepts |
|
|
26 | (3) |
|
AI Solutions: Image Analysis |
|
|
29 | (5) |
|
Machine Learning Impacts Several Medical Specialties |
|
|
34 | (12) |
|
AI and Medication Management |
|
|
46 | (5) |
|
|
|
51 | (1) |
|
|
|
52 | (7) |
|
Chapter 3 AI Criticisms, Obstacles, and Limitations |
|
|
59 | (12) |
|
Explainability Remains a Challenge |
|
|
59 | (5) |
|
Generalizability Remains Elusive |
|
|
64 | (2) |
|
Addressing Hype, Fraud, and Misinformation |
|
|
66 | (1) |
|
|
|
67 | (1) |
|
|
|
68 | (3) |
|
Chapter 4 CDS Systems: Past, Present, and Future |
|
|
71 | (24) |
|
CDS Has Improved Dramatically Over Time |
|
|
71 | (1) |
|
How Effective Are CDS Systems? |
|
|
72 | (2) |
|
Obstacles to CDS Implementation and Effectiveness |
|
|
74 | (4) |
|
Commercially Available CDS Systems |
|
|
78 | (9) |
|
Specialized CDS Tools Improve Radiological Reports/Analysis |
|
|
87 | (2) |
|
Non-Commercial CDS Resources |
|
|
89 | (1) |
|
CDS Hits a Psychosocial Roadblock |
|
|
90 | (1) |
|
|
|
91 | (4) |
|
Chapter 5 Reengineering Data Analytics |
|
|
95 | (10) |
|
The Future of Subgroup Analysis |
|
|
95 | (4) |
|
Predicting MS and Emergency Response |
|
|
99 | (1) |
|
Big Data Meets Medication Management |
|
|
100 | (1) |
|
The Role of Data Analytics in Cancer Risk Assessment |
|
|
101 | (1) |
|
Impact of Data Analytics on Healthcare Costs |
|
|
102 | (1) |
|
|
|
103 | (2) |
|
Chapter 6 Will Systems Biology Transform Clinical Decision Support? |
|
|
105 | (18) |
|
Redefining Health and Disease |
|
|
107 | (3) |
|
Is Systems Biology Ready for Prime Time Medicine? |
|
|
110 | (4) |
|
The Whole Is Greater than the Sum of Its Parts |
|
|
114 | (1) |
|
Network Medicine's Essential Components |
|
|
114 | (5) |
|
Systems Medicine Research |
|
|
119 | (1) |
|
|
|
120 | (3) |
|
Chapter 7 Precision Medicine |
|
|
123 | (16) |
|
Addressing Genetic Predisposition |
|
|
126 | (3) |
|
Pharmacogenomics: Precision Medicine's Low-Hanging Fruit |
|
|
129 | (4) |
|
The Future of Precision/Personalized Medicine |
|
|
133 | (2) |
|
|
|
135 | (4) |
|
Chapter 8 Reinventing Clinical Decision Support: Case Studies |
|
|
139 | (20) |
|
Improving Patient Scheduling, Optimizing ED Functioning |
|
|
143 | (3) |
|
Embracing Mobile Tools That Improve Clinical Decision Making |
|
|
146 | (7) |
|
Technological Approach to Diagnostic Error Detection |
|
|
153 | (2) |
|
Promising Solutions, Unrealistic Expectations |
|
|
155 | (2) |
|
|
|
157 | (2) |
| Index |
|
159 | |
Paul Cerrato, MA, has more than 30 years of experience working in healthcare as a medical journalist, research analyst, clinician, and educator. He has written extensively on clinical medicine, clinical decision support, electronic health records, protected health information security, and practice management. He has served as the Editor of InformationWeek Healthcare, Executive Editor of Contemporary OB/GYN, Senior Editor of RN Magazine, and contributing writer/editor for the Yale University School of Medicine, the American Academy of Pediatrics, InformationWeek, Medscape, Healthcare Finance News, IMedicalapps.com, and MedpageToday. The Health Information Management Systems Society (HIMSS) has listed Mr. Cerrato as one of the most influential columnists in healthcare IT. He has served as a guest lecturer or faculty member at the Columbia University College of Physicians and Surgeons, Harvard Medical School, and Vermont College. Among his achievements are 6 editorial awards from the American Business Mediaoften referred to as the Pulitzer Prize of business journalismand the Gold Award from the American Society of Healthcare Publications Editors for best signed editorial.
John D. Halamka, MD, MS, president of the Mayo Clinic Platform, leads a portfolio of new digital platform businesses focused on transforming health by leveraging artificial intelligence, machine learning, and an ecosystem of partners for the Mayo Clinic. He is a practicing emergency medicine physician. Previously, Dr. Halamka was executive director of the Health Technology Exploration Center for Beth Israel Lahey Health in Massachusetts. Previously, he was chief information officer at Beth Israel Deaconess Medical Center for more than 20 years. In addition, he was the International Healthcare Innovation Professor at Harvard Medical School. As the leader for innovation at the $7 billion Beth Israel Lahey Health, he oversaw digital health relationships with industry, academia, and government worldwide. As a Harvard Medical School professor, he served the George W. Bush administration, the Obama administration, and governments around the world planning their health care information (IT) strategies.
