Healthcare industry comfort with artificial intelligence and machine learning models – as well as the skill to deploy them in a multitude of clinical, financial and operational use cases – continues to increase in 2023.
There is growing evidence that training AI algorithms on multiple datasets can improve decision support, enhance population health management, streamline administrative tasks, allowing for cost savings and even improved results.
But much remains to be done to ensure results are accurate, reliable, understandable, and evidence-based for patient safety and health equity.
Undoubtedly, the application of AI in healthcare went far beyond “real life” in 2019 and was significantly invested by providers and payers last year. This year, we covered further industry discussions focused on beliefs and best practices. We’ve highlighted industry views on the merits of neural networks and deep learning and how to remove data barriers, along with an announcement of successful studies and, of course, relationships. new healthcare AI technology partner.
How AI bias occurs – and how to eliminate it. Although posted about 30 days before the end of 2021, readers flocked to the advice of Stanford cardiologist Dr. Sanjiv M. Narayan, co-director of the Stanford Arrhythmia Center and director of the Stanford Arrhythmia Center. Director of the Atrial Fibrillation Program and professor of medicine at Stanford University. of Medicine. Narayan discussed multiple approaches to removing bias in AI, including training multiple versions of the algorithm, adding multiple datasets to the AI, and updating the machine’s training dataset over time. He cautions that algorithmic sanitization strategies are not perfect, and that bias is more likely to compound when integrating complex systems.
Step by step growing trust in healthcare AI. While the use of AI in healthcare has increased, providers have been concerned about how much they should trust machine learning in clinical settings. A Chilmark Research report by analyst Dr Jody Ranck indicates that, based on a review of hundreds of COVID-19 pandemic algorithms in the first year, many AI cases cannot be validated. Ranck proposed strategies to enhance evidence-based AI development.
A sentient AI? Convincing you that it’s people is only part of LaMDA’s job. In this guest post, published after a frenzy in the mainstream media about a seemingly “sentient” machine learning application. Dr Chirag Shah, an associate professor at the University of Washington School of Information, explains how Google’s LaMDA chatbot, which easily passed the Turing Test, failed to demonstrate the presence of a sense of self-awareness. LaMDA just proves that it can create the illusion of possessing self-awareness – which is exactly what it was designed to do.
Duke, Mayo Clinic, others launch innovative AI collaboration. Artificial intelligence researchers and technology leaders from Duke, Mayo Clinic, University of California Berkeley, and others announced a new Health AI Partnership at a HIMSS virtual learning event today. before the end of 2021. By developing an online curriculum to help train IT leaders and work with stakeholders, the collaborators are aiming to develop a process-based standardized evidence for AI implementation in healthcare.
The intersection of remote patient monitoring and AI. Robin Farmanfarmaian, author of “How AI Can Democratize Healthcare: The Rise of Digital Care” and four other books, discussed how AI is impacting remote patient monitoring today and how it can democratize healthcare. “RPM has the potential to collect clinical-level data when people of all health stages and all ages,” she said. degrade EHR data from hospitals or health systems.”
Mayo Launches AI Startup Program, With Support From Epic and Google. In March, the Mayo Clinic kicked off a 20-week startup program to support early-stage health-tech AI companies. The clinic’s technology, health, and business experts, along with thought leaders from Google and Epic, will provide expertise to the team to help startups outline their futures. requirements of the AI model.
AI study finds 50% of patient notes duplicate. Researchers from the Perelman School of Medicine at the University of Pennsylvania in Philadelphia used natural language processing to find the rate of note duplication, as well as the annual rate of duplication, on records of 1, 96 million patients between 2015 and 2020. cast doubt on the authenticity of all information in their medical records, making it difficult to find and verify information in daily clinical work,” according to the report. Their JAMA report was published in September.
AWS, GE leaders discuss barriers to data sharing, AI implementation. In a sideline chat at HIMSS22, Dr. Taha Kass-Hout of Amazon Web Services and Vignesh Shetty of GE Healthcare discussed AI challenges and opportunities for better connected decision-making.
How AI and machine learning can predict disease and promote health equity. In a recent Q&A, Brett Furst, president of HHS Tech Group, discussed how to leverage the COVID-19 Research Database – one of the most comprehensive cross-linked datasets available. world – it is possible to establish a cause-and-effect relationship between many variables. When machine learning technology determines how many variables interact, it can reliably predict health outcomes.
CommonSpirit Health goes big with AI-integrated OR scheduler. This case study, starring Brian Dawson, vice president of perioperative services systems at CommonSpirit, shows how health systems implement AI-powered tools to improve operating room efficiency across the board. their 350 hospitals. “Healthcare providers globally have had to do more for less, and that has led to increased burnout, staff shortages, dissatisfaction,” said Dawson. of patients and scarce resources”.
Andrea Fox is the senior editor of Healthcare IT News.
Email: [email protected]
Healthcare IT News is a publication of HIMSS.
