As artificial intelligence becomes more common in healthcare systems, healthcare professionals must ask the right questions for AI to live up to expectations, according to a viewpoint article published in JAMA.
Thomas M. Maddox, MD, MSc, of the Washington University School of Medicine in St. Louis, Missouri, and colleagues, broadly define AI as a field of computer science that aims to replicate human intelligence through computer systems.
The first thing that the authors considered is what tasks AI is suited for in the healthcare realm. Although AI can be taught to identify clinically useful patterns in large, high-dimensional data sets, for data sets with less rigid, agreed-upon criterion, it may be difficult to train an AI algorithm successfully.
Currently, the authors argued, the data contained in electronic health records is not suitable to be used by AI. AI works best with high-quality data sources, while electronic health records and medical billing claims tend to be ill defined with broad, limited categories.
Want to publish your own articles on DistilINFO Publications?
Send us an email, we will get in touch with you.
Like any other medication or medical device, Dr Maddox and colleagues argued that new AI tasks must undergo extensive evaluation to determine their clinical efficacy and safety before they can be widely implemented. The current evidence standard for AI is not well defined; however, the researchers believe that it should be proportional to the task.
In addition to being clinically effective, AI must also be cost-effective to truly benefit the healthcare sphere.
To best integrate AI into health care, physicians need to learn how to adapt to AI-supported insights in a way that does not overwhelm clinical teams. The authors suggested that the best way for AI to work is alongside, rather than as a replacement for, human intelligence.
“Whether AI will ultimately improve quality of care at reasonable cost remains an unanswered, but critical, question,” they concluded. “Without the difficult work needed to address these issues, the medical community risks falling prey to the hype of AI and missing the realization of its potential.”
Date: January 15, 2019
Source: Medicalbag
