Generative AI boosts radiology productivity at Northwestern Medicine

Webp 586xm2q3cvywyzzsuz2is9bx9xzl
Luke Figora Vice President for Operations and Chief Operating Officer | Northwestern University

Generative AI boosts radiology productivity at Northwestern Medicine

A groundbreaking generative AI system developed by Northwestern Medicine is transforming radiology. The technology has demonstrated the potential to increase productivity, enhance accuracy, and address global shortages in the field of radiology. According to a study published in JAMA Network Open, the AI system boosts radiologists' productivity by an average of 15%, with some reaching up to 40% improvement.

"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," said Dr. Mozziyar Etemadi, senior author and assistant professor at Northwestern University Feinberg School of Medicine.

Deployed across the Northwestern Medicine network's 11 hospitals, nearly 24,000 radiology reports were analyzed over five months in 2024. The findings showed significant efficiency gains without sacrificing accuracy. Additionally, unpublished research indicates potential improvements up to 80%, extending capabilities to CT scans.

The AI tool is noted as being the first of its kind integrated into clinical workflows worldwide. It analyzes X-rays or CT scans comprehensively and generates personalized reports that are mostly complete before being reviewed by radiologists.

"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," stated Dr. Samir Abboud from Northwestern Medicine.

In addition to improving workflow efficiency, the system flags life-threatening conditions like pneumothorax immediately—before human review—allowing for faster triage in emergency settings.

"On any given day in the ER, we might have 100 images to review... This technology helps us triage faster," Abboud explained.

Dr. Etemadi emphasized that this approach does not rely on large tech companies: "There is no need for health systems to rely on tech giants."

Dr. Jonathan Huang added: "Our study shows that building custom AI models is well within reach of a typical health system."

As the U.S faces a projected shortage of up to 42,000 radiologists by 2033 due to rising imaging demands and slow growth in residency positions, this innovation offers promising relief without replacing human expertise entirely.

"You still need a radiologist as the gold standard," said Abboud.

Northwestern's proprietary technology has received two patents with more pending approval stages as commercialization efforts begin.

The study titled "Efficiency and Quality of Generative AI–Assisted Radiograph Reporting" outlines these advancements further.

Mentioned in this story

LETTER TO THE EDITOR

Have a concern or an opinion about this story? Click below to share your thoughts.
Send a Letter

Submit Your Story

Know of a story that needs to be covered? Pitch your story to The Southland Marquee.
Submit Your Story

Mentioned in this story

Northwestern University

More News