Industry experts advocate for radiologists to opt for relying on artificial intelligence – rather than non-physician providers – for help with workflow and cost reduction.
Despite recent evidence that physician extenders can streamline radiology workflow and reduce turnaround times, radiologists should rely more on artificial intelligence (AI) for assistance than non-physician providers (NPP), a group of industry experts has said.
In an editorial published in the Journal of the American College of Radiology, a team of experts, led by Daniel Ortiz, M.D., with Summit Radiology in Georgia, pointed to the several benefits of AI – not only do the tools save money and streamline workflow, but they will not encroach on a radiologist’s responsibilities.
For these reasons, they said, radiologists should forego giving physician extenders – the nurse practitioners, physician assistants, and other providers who take on some of a provider’s duties – full practice authority.
“Although labor costs have been reduced and radiologists can focus more on complex imaging studies and interventional procedures, there are unintended consequences of non-physician practitioners in practice that could diminish physician’s role as healthcare providers,” the group wrote. “Therefore, we encourage radiologists to consider an alternative to non-physician practitioners in radiology: the incorporation of rapidly evolving artificial intelligence algorithms into daily practice.”
Related Content: Radiology Extenders Outperform Radiology Residents with Chest X-ray Interpretations
Their concern was spawned by a recently passed Georgia law that allows advanced practice registered nurses to order CR, MRI, and other imaging exams under certain circumstances. Some pockets of support in healthcare have heralded the measure as a way to both decrease labor costs and take less complicated studies off of the radiologist’s work list, leaving him or her available for more complex exams.
But, according to Ortiz and his colleagues, radiology must be careful before proceeding down this path.
“If one year, we agreed upon a catalog of procedures to be relegated, do we really believe it will stop at those procedures deemed ‘simple’ by radiologists?” they asked. “Radiology, particularly interventional radiology, already finds itself in ever-growing struggles with other physicians for access to procedures, such as cardiac imaging or vascular interventions.”
Instead, in their editorial, they outlined their reasons why radiologists should lean more heavily on AI for support.
First, they pointed out that the push behind greater autonomy for NPPs is rooted in their ability to complete non-complex tasks, such as placing a nasogastric tube, successfully. However, radiology’s interpretation responsibilities are largely more complicated.
“The fallacy of this logic is highlighted by the idea that radiologists are not only responsible for answering the question at hand, but are responsible for the ‘whole image,’” they explained. “Although one could train an NPP to identify tips of lines, tubes, and otherwise, this does not rescind the liability for finding the unexpected.”
Even if radiology does cede control of simpler tasks to NPPs – but remains responsible for signing off on the work completed – the specialty runs the risk of creating a “culture of rubber-stamping” without review.
Second, the Center for Medicare & Medicaid Services recently granted its first approval to AI software, allowing for payment for the use of these tools. Consequently, many in the industry see this as an encouragement to additional AI vendors to continue development and clinical trials.
“Unlike humans with a narrow scope of practice, narrow AI algorithms can easily be integrated into existing workflow and amalgamated into a platform for use by radiologists,” the group said, noting there is little support for autonomously operating, unsupervised AI. “Many of the gains touted by proponents for the use of preserving the radiologist’s role, supplemented by AI.”
Ultimately, said Ortiz and his colleagues, expanding the use of NPPs will have unforeseen consequences for radiology’s future. Rather than opting to delegate radiology work to others, providers should be focused on strategies that can empower radiologists to stay in control of the imaging care that benefits patients.
“Empowered by AI,” the group said, “radiologists can achieve their wish of being as ubiquitous as water and remain heralded as diamonds.”
For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.
MRI-Based Deep Learning Algorithm Shows Comparable Detection of csPCa to Radiologists
May 8th 2024In a study involving over 1,000 visible prostate lesions on biparametric MRI, a deep learning algorithm detected 96 percent of clinically significant prostate cancer (csPCa) in comparison to a 98 percent detection rate for an expert genitourinary radiologist.
FDA Clears AI-Powered Qualitative Perfusion Mapping for Cone-Beam CT
May 6th 2024Reportedly validated in more than 10 clinical trials, the AngioFlow perfusion imaging software enables timely identification of brain regions with cerebral blood flow reduction and those with significant hypoperfusion.
The Reading Room Podcast: Emerging Trends in the Radiology Workforce
February 11th 2022Richard Duszak, MD, and Mina Makary, MD, discuss a number of issues, ranging from demographic trends and NPRPs to physician burnout and medical student recruitment, that figure to impact the radiology workforce now and in the near future.
New Literature Review Finds ChatGPT Effective in Radiology in 84 Percent of Studies
April 29th 2024While noting a variety of pitfalls with the chatbot ranging from hallucinations to improper citations, the review authors found the use of ChatGPT in radiology demonstrated “high performance” in 37 out of 44 studies.