MONDAY, June 9, 2025 (HealthDay News) -- Artificial intelligence (AI) can boost radiologists’ ability to quickly and accurately assess people’s X-rays, a new study says.
Some radiologists became 40% more efficient at reviewing images with the help of a custom-built AI program, researchers reported June 5 in JAMA Network Open.
“This is, to my knowledge, the first use of AI that demonstrably improves productivity, especially in health care. Even in other fields, I haven’t seen anything close to a 40% boost,” senior researcher Dr. Mozziyar Etemadi said in a news release. He's an assistant professor of anesthesiology and biomedical engineering at Northwestern University in Chicago.
Northwestern engineers built the AI system from scratch, using clinical data from the Northwestern Medicine network.
The AI system analyzes X-ray imaging and issues a report that is 95% complete and personalized to each patient, researchers said.
A radiologist then reviews the AI report along with the imaging scan and issues their own diagnosis.
The system also flags life-threatening conditions like a collapsed lung in real time, and an automated tool cross-checks the AI report against patient records. Radiologists are immediately alerted of new conditions that require urgent intervention.
For this study, the AI system was deployed in real-time across the 12-hospital Northwestern Medicine network.
Nearly 24,000 radiology reports were analyzed during five months of 2024, half with and half without AI assistance, researchers said.
Results found an average boost of nearly 16% in efficiently completing X-ray reports, without compromising accuracy. Some radiologists achieved efficiency gains as high as 40%.
“For me and my colleagues, it’s not an exaggeration to say that it doubled our efficiency. It’s such a tremendous advantage and force multiplier,” said researcher Dr. Samir Abboud, chief of emergency radiology at Northwestern Medicine.
“On any given day in the ER, we might have 100 images to review, and we don’t know which one holds a diagnosis that could save a life,” Abboud said. “This technology helps us triage faster — so we catch the most urgent cases sooner and get patients to treatment quicker.”
Researchers said the system has since been further improved to enable AI assessment of CT scans. These latest tweaks have shown up to 80% efficiency gains, according to as-yet-unpublished results.
The Northwestern team also is adapting its AI to spot potentially missed or delayed diagnoses for difficult-to-detect conditions like early-stage lung cancer.
“Our study shows that building custom AI models is well within reach of a typical health system, without reliance on expensive and opaque third-party tools like ChatGPT," Etemadi said. "We believe that this democratization of access to AI is the key to drive adoption worldwide.”
These results indicate that AI could help manage a growing shortage in radiology specialists.
By 2033, the U.S. is expected to be short as many as 42,000 radiologists, researchers said in background notes. Meanwhile, imaging is projected to rise by 5% annually while radiology positions increase by just 2%.
However, AI will not replace humans, Abboud said.
“You still need a radiologist as the gold standard,” he said. “Medicine changes constantly — new drugs, new devices, new diagnoses — and we have to make sure the AI keeps up. Our role becomes ensuring every interpretation is right for the patient.”
Northwestern has taken out two patents on the technology, and others are pending, researchers said.
More information
The Mayo Clinic has more on AI in health care.
SOURCE: Northwestern University, June 5, 2025