 |
. | Harvard researchers say your next ER diagnosis may come from AI—and it could be more accurate than human doctors Story by Amaya Nichole |
A new peer-reviewed study found AI diagnosed emergency patients more accurately than human doctors. Researchers believe it could reshape the future of medicine.
What if the most accurate diagnosis in the emergency room didn’t come from a doctor — but AI? A new study led by a team of researchers at Harvard Medical School suggests that moment may already be here.
The study, which was published in Science, a peer-reviewed journal, found that AI systems outperformed doctors in a number of emergency medicine situations. In diagnosing patients, the large language model (LLM) from OpenAI was more accurate than the human experts.
This revelation comes as AI is already gaining traction in medicine, with nearly one in five U.S. physicians now using AI to assist with diagnosing patients, according to research published last month by the American Medical Association.
“The rapid pace of improvement in LLMs has substantial implications for the science and practice of clinical medicine. Although applying AI to assist with clinical decision support is sometimes viewed as a high-risk endeavor, greater use of these tools might serve to mitigate the human and financial costs of diagnostic error, delay, and lack of access,” the study authors wrote.
In one experiment conducted by the researchers, 76 patients at a Boston emergency room were assessed by AI and a pair of human doctors. Both sides were given the same electronic health records containing a patient’s vital signs, demographic information, and a brief nurse’s note.
In the triage situations with limited information, AI identified the correct or near-correct diagnosis in 67 percent of cases, compared to 50 to 55 percent for the doctors. When more detail was available, the AI’s accuracy in diagnosing rose to 82 percent, compared to 70 to 79 percent for the doctors.
For the study, the researchers also gave five clinical summaries based on real-life cases to both the LLM and a group of 46 doctors using conventional resources. While AI received a median score of 89 percent for its long-term treatment plans, the team of experts earned a median score of 34 percent.
Despite the promising results, the researchers were careful to note that “our study addresses only text-based performance for both humans and machines; clinical medicine is multifaceted and awash with non-text inputs, including auditory (such as the patient’s level of distress) and visual information (for example, interpretation of medical imaging studies) that clinicians routinely use.”
Still, the researchers see the findings as a signal of what may be ahead. “I don’t think our findings mean that AI replaces doctors,” said Arjun Manrai, PhD, one of the study’s authors, who leads an AI lab at Harvard Medical School, according to The Guardian. “I think it does mean that we’re witnessing a really profound change in technology that will reshape medicine.”
Adam Rodman, MD, an author of the story, told The Guardian that he predicted AI would not replace physicians but merge with them in a new “triadic care model…the doctor, the patient, and an artificial intelligence system.”
Rodman also acknowledged one of the central questions regarding AI in clinical settings: responsibility. “There is not a formal framework right now for accountability,” Rodman told The Guardian, adding that patients ultimately “want humans to guide them through life or death decisions [and] to guide them through challenging treatment decisions.”
As AI continues to advance, the debate over its role in medicine may also shift, raising the question not of whether it belongs in the exam room—but the role that it will play in it.
No comments:
Post a Comment