Why are doctors hesitating to use ChatGPT in medicine?

1 of 1

New York. Artificial intelligence systems like ChatGPT are coming into everyday use. However, it is less likely to be adopted by doctors due to lack of skill to interpret and act accordingly. This has been revealed in a study.

Like every other industry, physicians will soon begin to incorporate AI tools into their clinical practice, helping them make critical decisions on the diagnosis and treatment of common medical conditions.

These tools, called clinical decision support (CDS) algorithms, could be helpful in guiding health care providers, for example, on which antibiotics to prescribe or whether to recommend risky heart surgery.

According to a new perspective article published in the New England Journal of Medicine, the success of these new technologies depends to a large extent on how physicians act. It requires a unique set of skills which many currently lack.

CDS algorithms can include everything from risk calculators to sophisticated machine learning and artificial intelligence-based systems.

They can be used to predict which patients are most likely to go on to life-threatening sepsis from uncontrolled infection or which therapy is most likely to prevent sudden death in a heart patient.

Daniel Morgan, Professor of Epidemiology and Public Health at the University of Maryland School of Medicine (UMSOM) and co-author of the Perspective, said, “These new technologies have the potential to make a significant impact on patient care, but doctors have yet to integrate them into their medical practice. Before incorporating algorithms, it has to be learned how machines think and work.

While some clinical decision support tools have already been incorporated into electronic medical record systems. Medical care providers often find the software difficult to use.

“Doctors don’t need to be math or computer experts, but they should have a baseline understanding of what probability and risk adjustment mean,” said Kathryn Goodman, Professor of Epidemiology and Public Health at UMSOM and co-author of Perspectives. What an algorithm does, but most have never been trained in those skills.

To address this gap, medical education and clinical training need to include explicit coverage specifically tailored to the CDS algorithms.

He also proposed that probabilistic skills should be learned early in medical schools, physicians should be taught to critically evaluate and use CDS predictions in their clinical decision making, and practice interpreting CDS predictions.

They must also learn to communicate with the patient about CDS-guided decision making.

(IANS)

read this also – Click to read the news of your state / city before the newspaper