With artificial intelligence poised to take over image-centric medical domains, some health experts are urging radiologists and pathologists to consider merging into a single specialty.
Advancements in deep learning have paved the way for computers to take on a bigger role in reading medical images, but that doesn’t mean machines will replace radiologists and pathologists entirely, according to a viewpoint published in JAMA. Instead, the authors argue that the two specialties should merge into a single role as “information specialists,” allowing computers to take over the menial tasks associated with reading images.
RELATED: RSNA16 - New technology, research and radiology careers
Artificial intelligence—including IBM’s Watson—can read thousands of X-rays and CT scans in a matter of minutes, identifying fractures or abnormalities far quicker than a radiologist. By adopting a modified role in order to make room for AI advancements, clinicians would oversee machine interpretations and make higher level clinical decisions by integrating information from patient charts.
“The merger is a natural fusion of human talent and artificial intelligence,” the authors wrote. “United, radiologists and pathologists can thrive with the rise of artificial intelligence.”
With @EricTopol. We ask radiology & path to merge into single specialty to manage AI to deliver better care https://t.co/oSzwvAKkxl #rsna16
— Saurabh Jha (@RogueRad) November 29, 2016
In an accompanying editorial, researchers at Harvard Medical School highlighted the rapid integration of AI in medicine over the last five years, pointing to recent advancements that could transform medical practices. Integrating validated models into image-based specialties allows hospitals to quickly evaluate millions of images at a fraction of the cost, freeing up clinicians to focus on other patient care concerns.
Machine learning has been shown to improve cost and quality in healthcare, help clinicians classify and treat diseases, and address patient satisfaction. But experts have also cautioned that while computer systems are adept at processing vast quantities of data, evaluating nuanced information could pose a significant challenge requiring human intervention.