If you’re wondering how artificial intelligence (AI) will affect your healthcare institution, company, or career, the Association for the Advancement of Medical Instrumentation (AAMI) has answers.
AAMI has recently released an in-depth report, “Artificial Intelligence, the Trust Issue,” that provides expert insights for product developers, regulators, standards developers, hospital systems, healthcare technology management (HTM) professionals, risk managers, and clinicians. The report is the first of a new AAMI series of Medical Device Safety in Focus reports that will explore hot topics in healthcare technology.
According to AAMI’s Joe Lewelling, vice president of industry, the digital report, which consists of an estimated 40 pages of content including informative videos, figures, and expert insights, will serve as “an introduction for people in healthcare who need to know more about AI but are not necessarily specialists in AI.”
“Artificial intelligence is something that the healthcare industry is going to be focusing on, and working with, and struggling with for the next decade or longer,” says Lewelling “This document is not going to make an expert in artificial intelligence, but it is going to prepare you to make important decisions.”
“Artificial Intelligence, the Trust Issue,” tells a compelling story about the opportunities and risks of AI in healthcare—and provides wide-ranging perspectives on efforts to balance risks and opportunities. The report is presented in three sections:
- The first section, Rehumanizing Healthcare with AI, sets the stage with a look at the potential of AI to remake healthcare—and what’s driving high hopes Many experts believe that AI could revolutionize healthcare delivery in terms of access, quality, and outcomes, including health equity. AI also could rehumanize healthcare by freeing up clinicians to focus on patient care and driving efficiencies that improve patient experiences.
If you’re still trying to wrap your head around what AI can do, this section is a good primer. There are many examples of the benefits that AI-enabled health technologies are already delivering. But the AI era is still in its infancy. Market indicators and physician adoption rates suggest the potential for exponential growth in coming years.
“When done right, AI has enormous potential,” notes report contributor Jesse Ehrenfeld, president-elect of the American Medical Association, professor of anesthesiology at the Medical College of Wisconsin, and co-chair of AAMI’s Artificial Intelligence Committee. “Digital medicine has enormous opportunity to improve health outcomes. There is tremendous enthusiasm about disruptive innovation as long as it’s clinically validated. Where we hear concerns, and I hear this all the time as around lack of transparency that will interfere with that trust, with understanding how these tools were designed and validated.”
- Inside AI’s ‘Black Box’ focuses on the myriad ways that AI-enabled health technologies could jeopardize patient health and safety, increase inequalities and inefficiencies, undermine trust in healthcare, and adversely impact the management of healthcare.
There are concerns about the Big Data that makes AI possible, including data quality, access, bias, selection, and control, as well as change management of AI systems and the safety and security of cloud services. The Black Box conundrum makes it hard to understand what happens in very complex, multilayered systems that are capable of learning from large data sets, which engenders distrust. AI also can fail in “weird ways” and make mistakes that no human would make. Cybersecurity is a massive—and massively increasing—concern.
Finally, the potential of AI in healthcare is also luring algorithm engineers to product development, but their inexperience with medical device development, standards, and regulations can pose risks. AI health technology also raises basic questions about financial risk to healthcare systems and providers.
“The proliferation of developers who are good at developing artificial intelligence systems, but who’ve never done it in a life-sciences setting, can cause problems in terms of safety, efficacy, and unintended consequences,” said contributor Scott Thiel, global head of regulatory policy and intelligence, Hologic, Inc. “Frankly it can also cause compliance issues out of ignorance of the regulatory requirements for these systems.”
- Balancing Opportunities and Risks. Given the lofty expectations for AI to transform healthcare and the broad scope of vulnerabilities and risks, it will take broad collaboration among a wider variety of stakeholders to ensure the quality, safety, and effectiveness of AI-enabled healthcare technology.
Managing risks encompasses standards and guidance for risk management, regulatory clarity, and resources for risk management and governance in healthcare facilities. The report includes practice guidance and an AI vendor checklist for HTM professionals. It also addresses an irony in the growth of AI health technology and the complex challenges associated with it: Everyone involved in developing, managing, and using it needs to bolster their knowledge and competencies.
“Communication buys grace,” adds Mike Powers, system director of healthcare technology management, Intermountain Healthcare, and a member of AAMI’s Artificial Intelligence Committee. “The more manufacturers, AI developers, and organizations that have experience with AI products communicate the risks and benefits of AI technology, the more informed additional parties can be to make informed decisions.”
At the root of the report is the insights of professionals who are invested in a future where AI is used to enrich, not burden, the healthcare space.
“We know that artificial intelligence is going to change the world. Those working with healthcare technology may be wondering where they fit in or what they can do about it. The best action you can take—right now—is to get informed,” says Gavin Stern, MPH, MS, editor in chief at AAMI and the association’s director of publications.
“That’s why we developed Medical Device Safety in Focus, a new kind of publication that provides an in-depth look at a specific healthcare technology topic. The first edition of MDSIF puts you in the room with the foremost experts working at the confluence of artificial intelligence and healthcare, from multiple fields and viewpoints. This is the AI primer you’ve been waiting for.”
The report is free for AAMI Members and available for purchase at aami.org/MDSIF-AI. Interested parties may read the executive summary for free at aami.org/MDSIF-AI-summary.