WASHINGTON. As artificial intelligence (AI) technology rapidly evolves, the question on many professionals’ minds is: Will AI replace human expertise, or will it enhance it? For radiologists, this question has been looming for a decade, as advanced algorithms promise to improve diagnostic accuracy and efficiency in interpreting medical scans.
AI’s potential in healthcare, particularly in radiology, is a subject of both optimism and caution. The technology can now analyze scans, suggest diagnoses, and even draft reports, offering unprecedented support to radiologists. “Some of the AI techniques are so good, frankly, I think we should be doing them now,” said Dr. Ronald Summers, a radiologist and AI researcher at the National Institutes of Health. Summers’ lab has developed AI programs to detect various conditions, yet these innovations have not seen widespread adoption due to what he describes as the “culture of medicine.”
Since the 1990s, radiologists have utilized computers to enhance imaging. However, modern AI can go further, interpreting images and providing diagnostic insights. Despite the FDA approving over 700 AI algorithms for medical use, with more than 75% in radiology, only about 2% of radiology practices have adopted these technologies.
Skepticism among radiologists stems from concerns about the real-world testing of AI, transparency regarding its operations, and the demographics of training datasets. “If we don’t know on what cases the AI was tested, or whether those cases are similar to the kinds of patients we see in our practice, there’s just a question in everyone’s mind as to whether these are going to work for us,” noted Dr. Curtis Langlotz, a radiologist at Stanford University.
Currently, all FDA-approved AI programs require human oversight. In Europe, however, regulators approved the first fully automated software for reviewing and reporting on chest X-rays in 2022. The urgency for such technology in Europe is high, with some hospitals facing significant backlogs due to a shortage of radiologists.
In the U.S., fully automated AI screening remains on the horizon, primarily due to radiologists’ hesitance to relinquish control to algorithms. “We try to tell them they’re overtreating people and they’re wasting a ton of time and resources,” said Chad McClennan, CEO of Koios Medical, which provides AI tools for ultrasound imaging.
AI’s promise extends to enhancing accuracy and reducing costs. Research indicates that radiologists often disagree on diagnoses, with one study finding a 30% disagreement rate among physicians reviewing the same breast scans. About 20% of breast cancers are missed during routine mammograms, underscoring the potential impact of AI in improving detection rates.
In the short term, experts anticipate AI will function like an autopilot system in aviation, performing crucial tasks but always under human supervision. “I will tell patients, ‘I looked at it, and the computer looked at it, and we both agree,’” said Dr. Laurie Margolies of Mount Sinai hospital network, illustrating how AI can provide additional reassurance to patients.
Preliminary results from a Swedish study involving 80,000 women showed that a single radiologist working with AI detected 20% more cancers than two radiologists working without AI. This approach also reduced the workload by 44%, addressing the workforce shortage in breast radiology.
Despite these promising developments, the role of radiologists remains critical. “If an automated algorithm misses a cancer, that’s going to be very negative for trust in the caregiver,” warned Dr. Kristina Lang of Lund University, emphasizing the importance of human oversight.
Radiologists continue to double-check AI outputs to avoid liability issues, potentially diminishing the benefits of reduced workload and burnout. “Only an extremely accurate, reliable algorithm would allow radiologists to truly step away from the process,” said Dr. Saurabh Jha of the University of Pennsylvania.
As AI technology advances, the balance between human expertise and machine assistance will define the future of radiology, potentially transforming it from a field at risk of automation to one enhanced by the precise and tireless capabilities of AI.
Si Venus L Peñaflor ay naging editor-in-chief ng Newsworld, isang lokal na pahayagan ng Laguna. Publisher din siya ng Daystar Gazette at Tutubi News Magazine. Siya ay isa ring pintor at doll face designer ng Ninay Dolls, ang unang Manikang Pilipino. Kasali siya sa DesignCrowd sa rank na #305 sa 640,000 graphic designers sa buong daigdig. Kasama din siya sa unang Local TV Broadcast sa Laguna na Beyond Manila. Aktibong kasapi siya ng San Pablo Jaycees Senate bilang isang JCI Senator.
