CITATION: Agyemang-Gyau, P. (2021). Artificial Intelligence in Healthcare and the Implications for Providers. Online Journal of Nursing Informatics (OJNI), 25(2). https://www.himss.org/resources/online-journal-nursing-informatics
We are in the era of the Internet of things and Artificial intelligence (AI) continues to define this century. Artificial intelligence refers to a broad term that includes machine learning, natural language processing, rule based expert systems, physical robots, and robotic automation. The rise of computerized systems and medical devices in safely and efficiently diagnosing, treating, and coordinating care is a clear indication that AI is here to stay and grow in importance. While AI holds a lot of potential, the implications for primary care providers needs to be addressed as it may limit implementation. The application of AI in healthcare is also discussed.
When the concept of the amphibious car was announced a little over half a century ago, it was met with incredible delight, and there was much enthusiasm for the product. The idea of a car with the viability to ply roads and navigate water bodies was something to behold. Although this piece of ingenuity briefly grabbed traction, the amphibious car failed to grab market share sizeable enough to justify its existence. Eventually, the amphibious car fizzled into oblivion and into the dustbins of history. Since then, humanity has taken giant strides in technological advancements and although there are no flying cars in the 21st century, there has been a concerted effort to use technology to enhance everyday life. Reducing costly human errors has become essential in ensuring safety, efficiency, and productivity.
For instance, 94% of motor vehicular accidents (MVA) are a result of human error (National Highway Traffic Safety Administration [NHTSA], 2015). Hence, billions of dollars have been poured into the development of enhanced motor vehicular safety features that aid in mitigating the senseless loss of life from MVA. The solution is automation and technology companies in Silicon Valley such as Tesla, Waymo and Embark are working on producing fully autonomous vehicles equipped with autopilot features that include active and passive safety systems. The NHTSA acknowledges the benefits of automation and its potential to “remove human error from the crash equation” (U.S. Department of Transportation, 2019). The benefits remain enormous and the large-scale adaptation to automated vehicles has a lot of economic benefits. According to an NHTSA study in 2010, MVA crashes cost $242 billion in economic activity (National Highway Traffic Safety Administration, 2010). This includes a loss of $57.6 billion in productivity and an estimated $594 billion related to loss of human life and the associated decrease in quality of life (National Highway Traffic Safety Administration, 2010). The money and resources saved can certainly be channeled towards more productive endeavors.
Similarly, the cost of human-induced errors and waste in the U.S medical system cannot be ignored. According to research funded by the Agency for Healthcare Research and Quality (AHRQ, 2014), approximately 12 million people are affected by diagnostic errors annually. Additionally, the cost of healthcare related waste in the U.S amounts to approximately $960 billion.
Decades ago, the Institute of Medicine (IOM) released a groundbreaking report that shed light on human-inspired medical errors in the American healthcare system. Its report entitled “To Err is Human: Building a Safer Health System,” identified the limitations of the U.S healthcare system and opened the door for improvements. According to the IOM report, “an estimated 44,000 to 98,000 people die annually from medical errors” (Wakefield, 2000, p. 233). These numbers are astonishing and indicate the need for intervention in an ever-changing healthcare system.
To put things in perspective, “more people die in a given year as a result of medical errors than motor vehicle accidents, breast cancer or AIDS” (Wakefield, 2000, p. 233). Since then, policies have been instituted and there has been a concerted effort from stakeholders, including the government and private entities, to remedy the situation. Nevertheless, medical errors continue to be the third leading cause of death in the United States and accounts for ten percent of all mortalities in the country.
Additionally, the cost of healthcare remains high and 28.5 million Americans do not have access to healthcare (United States Census Bureau, 2018). Furthermore, attention needs to be steered away from merely treating acute illnesses and focused on preventative medicine that improves quality of life and keeps individuals and families out of the hospital. Solutions are plentiful and the recent global health crisis has shed light on the importance of investing heavily in technology.
Telehealth has emerged as an essential tool in remotely diagnosing, monitoring, and triaging patients for COVID-19 at a fraction of what an emergency room visit may cost. And this can be done conveniently from the confines of the home environment without making the dreaded visit to the hospital. According to Goldman Sachs, Telehealth and other related digital health technology has the capacity to save approximately $300 billion in costs related to the management of certain chronic conditions (Stern, 2015). This innovation is inspired by Artificial Intelligence and in the era of Alexa and Google Home, it is refreshing to envision what the future has in store for healthcare. But what is Artificial intelligence?
Artificial intelligence (AI) refers to “a wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence” (Christopher, 2020, p. 1). We are in the era of the Internet of things (IoT) and Artificial intelligence may define this century. The term Artificial intelligence was first coined in the 1940’s and represents a broad term that includes machine learning, natural language processing, rule based expert systems, physical robots and robotic automation.
Machine learning which is a subset of artificial intelligence refers “to a process in which computers use algorithms to analyze large data sets in non-linear ways, identify patterns, and make predictions that can be tested and confirmed” (General Electric, 2016, p. 1). This form of technology is what makes it possible to stream TV shows on Netflix while commanding Siri to provide you with the most current weather report. This process focuses on building computer algorithms that can methodically learn, improve, and make decisions through experience without necessarily being programmed.
Furthermore, machine learning has key elements which include deep learning. Deep learning utilizes networks that are built on the premise of the neural network and neural activities in the brain. In 1943, McCulloch and Pitts took the initiative and began the early stages of the development of the perceptron, which represents the most fundamental unit of deep neural networks (Kawaguchi, 2017). The perceptron in its formative stages was reported in the New York Times in 1958 as “the embryo” of the electronic computer and a device that “will be able to walk, talk, see, write, reproduce itself and be conscious of its existence” (New York Times, 1958, p. 25). By imitating the functioning capacity of the human brain, it utilizes artificial neural networks to learn and develop solutions to complex and challenging problems that are beyond the limitations of the human brain. In brief, “Artificial intelligence is a set of algorithms and intelligence to try to mimic human intelligence. Machine learning is one of them, and deep learning is one of those machine learning techniques” (Christopher, 2020, p. 1). This piece of ingenuity is present in various forms of technology and brings into reality ideas such as robotic surgery.
There is a plethora of instances of this level of technology in healthcare. An example is the federally funded Undiagnosed Disease Network, that operates with private medical universities like Harvard and Stanford University to diagnose and treat rare diseases. Using data analysis, deep learning and genetic sequencing, the program has successfully diagnosed 25% of cases and provided a lifeline for many patients. At Stanford University, an algorithm was developed to accurately predict the prognoses of cancer patients that researchers suggest “could provide rapid and objective survival prediction for numerous patients” (Koontz, 2017). Developers at Google Health utilized deep learning to create a model capable of diagnosing breast cancer after analyzing an enormous amount of data from the United States and the United Kingdom. This study involved information gathered from over 28,000 women across both countries. Surprisingly, the system was able to learn and detect breast cancer with 5.7% less false-positives and 9.4% less false-negatives than board-certified radiologists (Abbasi, 2020). Based on the outcomes indicated, fully integrating this system into clinical practice has enormous potential in reducing misdiagnosis and medical errors as most of the breast cancers identified by the Artificial intelligence model in the Stanford University study were invasive. In areas with poorly equipped health care systems, AI can play an important role in bringing affordable healthcare to the doorsteps of individuals. Today, smart phones have the capacity to be equipped with electrocardiogram and ultrasound functionalities that can be utilized in impoverished areas for diagnostic purposes. However, what are the implications for medical providers?
With so much optimism, it is easy to imagine a future where machines will completely replace providers and our hospitals equipped with devices that can diagnose and recommend treatments within hours or maybe minutes. While a fully automated healthcare future may not be available yet, it is imperative to recognize the essence of embracing artificial intelligence as an important member of the healthcare interdisciplinary team. This important tool will be useful in streamlining processes so that providers can function more efficiently and effectively. As healthcare costs are reduced, clinicians will be afforded valuable time to be present for the patient.
Providers have expressed considerable skepticism about the potential scope of Artificial Intelligence “with only a few considering changes to current practices likely” (Blease et al., 2019, p. 13). There are grave implications if providers are not equipped with the educational curricula necessary to spearhead the incorporation of Artificial Intelligence in the healthcare system. “Improvements in education, may go some way in closing the rift between current AI health researchers and practitioners” (Blease et al., 2019, p. 13).
Additionally, healthcare providers need to adapt in a dynamic healthcare system to harness the full practicability of Artificial Intelligence. It is important to allay the fears of providers that are slow to adjust and concerned about the prospects of artificial intelligence completely replacing them. Krittanawong (2017) is emphatic in indicating that AI “cannot engage in high-level conversation or interaction with patients to gain their trust, reassure them, or express empathy, all important parts of the provider–patient relationship” (p. 2663). For now, providers who are uncertain about the prospects of Artificial Intelligence taking over the role of their practice can be at ease and reap the benefits of a harmonious relationship that can only benefit the patient.
Our world is evolving, and the novel coronavirus continues to alter the future of healthcare as we know it. The post-coronavirus era provides an opportunity to focus on closing the healthcare disparities that exist and limiting the human-factor in medical errors. It provides an opportunity to fully harness the potential of artificial intelligence to solve complex medical mysteries that plague humanity and increase the efficiency of breast cancer screening. It provides an avenue to fully embrace artificial intelligence as an integral part of the healthcare team.
Powered by the HIMSS Foundation and the HIMSS Nursing Informatics Community, the Online Journal of Nursing Informatics is a free, international, peer reviewed publication that is published three times a year and supports all functional areas of nursing informatics.
Abbasi, J. (2020). Artificial Intelligence Improves Breast Cancer Screening in Study. JAMA:Journal of the American Medical Association, 323(6), 499.
Agency for Healthcare Research and Quality (AHRQ). (2014). Diagnostic Errors. PSNet: Patient Safety Network. https://psnet.ahrq.gov/perspective/diagnostic-errors
Blease, C., Kaptchuk, T. J., Bernstein, M. H., Mandl, K. D., Halamka, J. D., & DesRoches, C. M. (2019). Artificial Intelligence and the Future of Primary Care: Exploratory Qualitative Study of UK General Practitioners' Views. Journal of medical Internet research, 21(3), e12802. https://doi.org/10.2196/12802
Christopher. (2020). What is Artificial Intelligence? How Does AI Work? Medium. https://medium.com/technology-innovations-insights/what-is-artificial-intelligence-how-does-ai-work-585f2d2fdde5
General Electric. (2016). Big Data, Analytics & Artificial Intelligence. GE Healthcare. https://www.gehealthcare.com/static/pulse/uploads/2016/12/GE-Healthcare-White-Paper_FINAL.pdf
Kawaguchi, K. (2017, June 17). The McCulloch-Pitts Model of Neuron. The University of Texas at El Paso. http://wwwold.ece.utep.edu/research/webfuzzy/docs/kk-thesis/kk-thesis-html/node12.html
Koontz, C. (2017, January 4). Doctors, data, and diseases: How AI is transforming health care. Venture Beat. https://venturebeat.com/2017/01/04/doctors-data-and-diseases-how-ai-is-transforming-health-care/
Krittanawong, C., Zhang, H., Wang, Z., Aydar, M., & Kitai, T. (2017). Artificial Intelligence in Precision Cardiovascular Medicine. Journal of the American College of Cardiology, 69(21), 2657–2664. https://doi.org/10.1016/j.jacc.2017.03.571
National Highway Traffic Safety Administration. (2010). Automated Vehicles for Safety. United States Department of Transportation. https://www.nhtsa.gov/technology-innovation/automated-vehicles-safety
National Highway Traffic Safety Administration. (2015). Traffic Safety Facts. National Center for Statistics and Analysis. https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812115
New York Times. (1958, July 8). New Navy device learns by doing: Psychologist Shows Embryo of Computer Designed to Read and Grow Wiser. New York Times, 25. https://doi.org/https://www.nytimes.com/1958/07/08/archives/new-navy-device-learns-by-doing-psychologist-shows-embryo-of.html
Stern, C. (2015, June 15). Goldman Sachs says a digital healthcare revolution is coming — and it could save America $300 billion. Business Insider. https://www.businessinsider.com/goldman-digital-healthcare-is-coming-2015-6
U.S. Department of Transportation. (2019, September 14). A Vision for Safety: Safer Roads through Vehicle Automation. https://www.transportation.gov/bts/bts/comment/reply/285821
United States Census Bureau. (2018, September 12). Health Insurance Coverage in the United States: 2017. https://www.census.gov/library/publications/2018/demo/p60-264.html
Wakefield, M., & DeLeon, P. H. (2000). To Err Is Human: An Institute of MedicineReport. Professional Psychology: Research & Practice, 31(3), 243-244. https://doi.org/10.1037/h0092814
The author is a native of Ghana and earned a full scholarship to further his education at Jacksonville University in Jacksonville, Florida. There, he served as the Vice-President of the Student Government in his senior year and graduated with University Honors, Departmental Honors and Magna Cum Laude as he was awarded a Bachelor of Science Degree in Nursing in 2015. After graduation, the author gained valuable experience as a telemetry nurse and served as the Chairman of the Nurse Research Council at St. Vincent’s Medical Center – Riverside in 2016. The primary author is a current Intensive Care Unit nurse at St. Vincent’s Medical Center – Clay and a graduate Doctor of Nurse Practice – Family Nurse Practitioner student at the University of North Florida.