According to Dr. David Classen, professor of medicine at the University of Utah and chief medical information officer at Pascal Metrics, patient safety is the one of the more pressing issues in the American medical system. But with the help of new tools like artificial intelligence (AI) and the adoption of new industry standards, there may be light at the end of the proverbial tunnel.
In a recent conversation with AAMI President and CEO Pamela Arora, Classen stated that patient safety has been top-of-mind for decades. The COVID-19 pandemic rolled back recent progress and the “latest studies suggest we’re still injuring one in four patients in hospitals.” However, Classen found reasons for optimism during the Augmented Ingenuity: AI for Patient Safety conference held in Salt Lake City, Utah. AI and other forms of advanced analytics can be leveraged to improve patient safety and prevent healthcare-associated infections.
The event’s attendees were diverse, with Classen describing the in-person collaboration as “intentionally cross-cutting.” It brought together various associations, health care providers, regulators like CDC and FDA, researchers, vendors and patient advocates. According to Classen and Arora, the event demonstrated the real need for the role of organizations like AAMI in the development, adoption and application of standards for AI-enabled medical devices.
First, Classen stated that there is a clear need for industry standards related to information technology (IT) and software that apply to AI. Unlike in the relatively nuanced case of electronic health records (EHR), there is notable public appetite for the development of AI standards.
Second, Classen indicated that AAMI is very well suited to fill this need. AAMI’s long history of standards development and engagement with important stakeholders means that the organization is “very well attuned to being able to create standards that are relevant here and to involve all the necessary stakeholders.” And AAMI’s reach is important, since “any company around the world that is creating any sort of regulated device is involved with AAMI.”
Last, Arora turned to the ANSI/AAMI HIT-1000 series of standards, which provide a framework for managing the safety and effectiveness of health IT software and systems. Classen co-chairs the committee responsible for ANSI/AAMI HIT-1000 alongside Jesse Ehrenfeld, president of the American Medical Association. Of note, he believes the standards series has the potential to springboard the development of new standards of AI because of their applicability to EHRs. Previously, EHRs were not subject to software standards that both EHR users and vendors could leverage. The new standards also accommodate for the fact that many EHR users customize vendor software. The solution? A novel set of standards that operate on a “shared responsibility” model to guide vendors and users as they test and utilize EHRs in different situations. Both Classen and Arora agreed that standards for AI would require a similar approach, as AI-enabled tools are likely to be adopted in site-specific ways.
Fortunately, there has already been some movement on the AI standards front for the medical field. AAMI’s own AI standards development committee remains hard at work developing the next in a series of standards dedicated specifically to the application of artificial intelligence in medical devices. In 2023, AAMI and BSI jointly published technical Information Report (TIR)34971:2023, Application of ISO 14971 to machine learning in artificial intelligence – Guide for the U.S. and U.K. The first-of its-kind document derives from a 2022 AAMI consensus report (CR) that the U.S. Food and Drug Administration recognized as “appropriate for meeting requirements for medical devices under the Federal Food, Drug, and Cosmetic Act.” In the near future, AAMI and BSI are planning to propose TIR34971 as an international standard through International Organization for Standardization (ISO) Technical Committee (TC) 210, Quality management and corresponding general aspects for products with a health purpose including medical devices, and IEC (International Electrotechnical Commission) Subcommittee (SC) 62A/Common aspects of medical equipment, software, and systems.
Additionally, the Augmented Ingenuity: AI for Patient Safety conference will return in the future. The latest iteration was generously funded by Robert Wood Johnson, but future funding will be provided by institutions including the University of Utah, the University of California San Diego, Carnegie Mellon University, and University of Pittsburgh.
