Artificial intelligence (AI) systems can provide vital support in clinical practice, but further improvements are needed, a new study has demonstrated.
Researchers from the University of Minnesota Medical School, Stanford University, Beth Israel Deaconess Medical Center and the University of Virginia have highlighted the pros and cons of diagnosing patients with GPT-4 – an AI large language model system.
During the US-based study, the team of scientists worked with 50 physicians in family medicine, emergency medicine and internal medicine.
- Research into genetic variations in people with IBS could lead to tailored diets
- Dr Sheila Reith: Insulin pen pioneer honoured at Pride of Britain Awards
- AI chatbots can give medical advice that could lead to death
They discovered that GPT-4 did not drastically enhance clinical care in comparison to conventional resources.
According to the results, GPT-4 worked better alone rather than when it was being used alongside clinicians using conventional diagnostic resources.
First author Dr Andrew Ol said: “The field of AI is expanding rapidly and impacting our lives inside and outside of medicine.
“It is important that we study these tools and understand how we best use them to improve the care we provide as well as the experience of providing it.”
- AI as effective as doctors for giving advice on back pain
- Patients prefer ChatGPT responses over doctors’ responses
- Nearly a quarter of GPs use AI to diagnose patients
He added: “This study suggests that there are opportunities for further improvement in physician-AI collaboration in clinical practice.”
Integrating AI into clinical practice is complex, according to health experts. However, there is potential for AI in healthcare, highlighting the importance of further exploration into how AI can best support clinical practice.
“Additional research is required to determine how clinicians should be trained to use these tools.
Read the study in full in the journal JAMA Network Open.