By Owen Quigley
One of the most important and largely invisible contributions that AI can make is to capture physician notes automatically. Medical professionals close to burnout do not benefit from manually entering data into electronic health records (EHRs).
Research published in EHR Intelligence recently found that 13% of stress and burnout self-reported by physicians were directly related to EHRs. The director of Biomedical Informatics Research at UNM, Philip Kroth, MD, asserts that clinician stress is largely caused by the design and structure of the clinical process, both of which correlate with EHRs.
As more physicians leave the profession, Kroth notes that the time spent documenting their work has resulted in a loss of quality time spent with their patients and families. "Medicals are finding that the goals of a traditional medical record have been hijacked in many ways."
UNM researchers worked with Stanford University, University of Minnesota, Hennepin County Medical Center, and Centura Health System in Colorado and Texas to survey 282 clinicians on the impact of EHR completion on stress and burnout.
Since EHRs were introduced ten years ago, medical record-keeping time has doubled
In 2009, the Health Information Technology for Economic and Clinical Health (HITECH) Act was signed into law. The survey shows that ten years after it was conducted, time dedicated to medical record-keeping doubled. For every minute physicians spend with their patients, they spend two minutes at the computer. Physicians' workdays have extended into their personal lives.
"Often, it takes a 60-hour week to keep up with documentation, and that's hard on family relationships and relationships," Kroth said. Artificial intelligence solutions can ease the burden by automating the process.
Real value is derived from data gathered from a doctor's conversation with a patient, or from a patient's case notes. The University of Florida's Clinical and Translational Science Institute director, Duane A. Mitchell, stated in the Axios article that AI products can take information and contextualize it so doctors can act on it.
In clinical trials, the time needed to identify the right set of patients is one example of value. According to Mona Flores, Global Head of Medical AI at Nvidia, the manual process takes weeks to extract data from databases, while AI models can do the work "within minutes".
Nvidia Partners with the University of Florida to Produce the GaterTron Model
Researchers from the University of Florida's academic health center announced on April 8 that they had collaborated with Nvidia to create GatorTron, an AI natural language processing model that can extract insights from massive amounts of clinical data very quickly.
According to a press release on the UFHealth website, the GatorTron Language Model is the first step toward a $100 million public-private partnership announced last year by the University of Florida and NVIDIA to pursue research in AI and supercomputing.
A variety of medical specialties, including oncology, internal medicine, and critical care, were analyzed in order to create GatorTron, using anonymous data from UF Health for 10 years. As part of the development of GatorTron, the University of Florida Institutional Review Board and the University of Florida Health Information Technology Center approved security controls to protect patients' privacy.
GatorTron represents the type of discovery that can happen when experts from academia and industry collaborate with leading-edge AI and computing resources," says David R. Nelson, M.D., senior vice president for health affairs at UF and president of UF Health. “Our partnership with Nvidia is crucial to UF emerging as a destination for AI expertise and development in health research.”
Other recent AI and healthcare news includes Microsoft's acquisition of Nuance Communications, a developer of voice recognition software technology with products that transcribe and analyze voice conversations between doctors and patients.
Last year, Nuance announced Dragon Ambient eXperience (DAX), a product that works in conjunction with EHR systems to capture and contextualize physician-patient conversations.
It is said that Nuance DAX enables physicians to create a voice-enabled exam room environment by leveraging Dragon Medical, which is used by over half a million physicians worldwide. Microsoft's cloud capabilities were incorporated into Nuance's offering.
Catholic non-profit integrated health system, SSM Health, plans to pilot this technology in some of its specialty clinics in St. Louis, Mo., Oklahoma, and Wisconsin. SSM Health Vice President and Chief Medical Information Officer, Ann Cappellari, MD, stated that the Nuance Dragon Ambient eXperience solution would allow providers to spend more time with their patients and less time on administrative tasks. “This facilitates better communication between patients and providers.”
The Mayo Clinic's Remote Device Management Program connects remote devices with AI services
On April 14, the nonprofit Mayo Clinic in Rochester, Minn. launched the Remote Diagnostic and Management Platform (RDMP), which is designed to help healthcare providers improve their use of connected health devices for monitoring patients remotely.
The RDMP connects devices to AI resources that can support clinical decision support and diagnoses, enabling what the Minnesota-based health system calls "event-driven medicine," according to an account in mHealth Intelligence.
AI will also change the way mental health services are delivered. According to Axios, "clinical psychiatry works much like it did 100 years ago, where a clinician sits down with a patient and, based on that discussion, creates a treatment plan," wrote psychiatrist Daniel Barron in his forthcoming book, "Reading Our Minds: The Rise of Big Data Psychiatry."
Instead, Barron envisions a future where those conversations can be recorded by AI models that analyze patients' speech and even facial expressions to detect mental illness and how to treat it. Of course, this would require patients to be comfortable with the idea that the AI would listen to and analyze the conversation with their therapist.
You can find the original articles and information at Axios, from EHR Intelligence, from a press release on the UFHealth website, from Nuance, and from mHealth Intelligence.