© Have a nice day – Stock.adobe.com
Artificial intelligence (AI) could revolutionize healthcare, so regulation is critical as more computers interact more with doctors and patients, said the U.S. Food and Drug Administration (FDA) commissioner. rice field.
Robert M. Calif, M.D., MACC, director of the FDA’s Food and Drug Administration, announced this month, “Patient Engagement in the Digital Health Era,” as part of the National Health Council’s 2023 Science Symposium for Patient Engagement in Washington, D.C. He delivered a lecture entitled “Empowerment”. , he said, dealt with “a range of exciting ‘hot’ topics” including AI.
Dr. Robert M. Caliph, MACC
U.S. Food and Drug Administration
FDA and Patients
Through projects such as patient-centered drug development initiatives and patient representation programs, the FDA has experience in integrating the voice of the patient into government work. These specifically add patient and caregiver perspectives, including inviting patients to FDA advisory committees and presenting at FDA meetings and workshops, Calif said.
Cooperation with AI
As for technology integration, Caliph said there is “great potential” for researchers entering the Fourth Industrial Revolution with rich and diverse digital data sources available in real time.
These include AI “large scale language models” that generate words and text based on huge databases and parameters.
“Large-scale language models are the next step that could usher in the revolution that many of us have been waiting for,” he said.
The ultimate goal of collecting data and evidence is to inform doctors and patients about the benefits and risks of medical interventions, improving health relative to the risks of harm, Khalif said.
“If you’re not agile with the use of large language models and regulations, you’ll quickly be overwhelmed by what you barely understand,” says Caliph. “Great things are really great. Imagine a world where your questions are answered instantly in a language that is appropriate for your literacy and numeracy. We can actually talk to you without having to spend all our time on .I could go on and on, but we believe regulation of large-scale language models is important to our future.”
Patients, technology, society
More data drives the development of medical products and treatments. This has important implications for food safety, tobacco products, and all FDA-regulated medicines, Calif said.
And there are other problems at the intersection of medicine and technology.
- Disparity. Life expectancy in the United States is declining, partly because of COVID-19, but also because of gun violence, suicide, and opioid overdoses. There are clear differences depending on the situation.
- Representativeness. The prevalence of mobile phones shows that technology is “embedded in human interactions,” Khalif said. Researchers now need to measure the diversity of values, preferences, beliefs, living circumstances, and economic factors.
- Algorithmic bias. Once used, algorithmic computer programs evolve and require updates and adjustments.
- Misinformation. People connect through social media networks, but AI gives “nearly everyone” the potential to create false narratives using fabricated images and sounds.
While private industry is looking for new ways to introduce machine learning and mathematical algorithms into everyday life, government regulation has lagged behind, Khalif said.
“Simply put, we need to mobilize resources to put these policies and tools in place and adaptively adjust our digital health efforts to support public health and regulatory innovation in a rapidly changing world. There is,” he said. “The potential certainly exists, and there is good reason to hope that innovative solutions will emerge that will realize the enormous potential of digital health technologies and lead to improved outcomes, but the risks also pre-exist. Better avoid it.”