Physicians’ lack of understanding is hindering the adoption of artificial intelligence (AI) tools in clinical practice, according to a recent perspective article published in the New England Journal of Medicine. While the integration of AI tools, such as clinical decision support (CDS) algorithms, has the potential to revolutionize medical decision-making, it heavily relies on physicians’ comprehension of these technologies.
AI systems, like ChatGPT, are increasingly being utilized in everyday practice to assist physicians in diagnosing and treating patients. CDS algorithms guide healthcare providers in critical determinations, ranging from prescribing antibiotics to recommending complex surgeries. However, the effective use of these tools depends on how well physicians interpret and act upon the risk predictions provided by the algorithms.
The authors of the article, faculty members from the University of Maryland School of Medicine, emphasize the need for physicians to enhance their understanding of AI tools to optimize their use in clinical practice. They argue that targeted training and a hands-on learning approach are essential for physicians to effectively employ these technologies.
Currently, many physicians face challenges in incorporating clinical decision support tools into their workflow, finding them cumbersome and difficult to use. Despite lacking expertise in mathematics or computer science, physicians need a baseline understanding of algorithms, probability, and risk adjustment to make the most of AI tools.
To bridge this gap, the authors suggest that medical education and clinical training should specifically address probabilistic reasoning tailored to CDS algorithms. They propose the establishment of the Institute for Health Computing (IHC) in collaboration with the University of Maryland. The IHC aims to leverage recent advancements in AI, network medicine, and other computing methods to create a premier learning healthcare system. The institute will offer education and training opportunities to healthcare providers on the latest technologies, including a certification in health data science.
Improving physicians’ probabilistic skills goes beyond the use of AI tools. Probability and risk analysis are fundamental to evidence-based medicine, and enhancing these skills can benefit patient care more broadly. The authors believe that the transformative era of medicine necessitates integration of vast amounts of data into machine learning systems, ultimately personalizing care for individual patients.
In conclusion, the successful adoption of AI tools in clinical practice hinges on physicians’ understanding of these technologies. By enhancing their comprehension through targeted training and a hands-on learning approach, physicians can effectively incorporate AI tools into their decision-making process. The Institute for Health Computing is poised to play a crucial role in educating and training healthcare providers on the use of AI in medicine. The future of healthcare lies in leveraging AI to deliver personalized care to patients, making it imperative for physicians to embrace and comprehend these technologies.
