Advancing Radiology with Continuous, Ambient, Low-Friction AI Technology

The principles of continuous, ambient, low-friction (CALM) technology focus on minimizing interruptions and enhancing communication by providing background support only when necessary. CALM AI is designed to integrate effortlessly into Picture Archiving and Communication Systems (PACS) and reporting systems, with the objective of alleviating cognitive load without introducing additional steps or alerts.

According to Shlomit Goldberg-Stein, MD, a professor of radiology at the Zucker-Hofstra School of Medicine and director of AI in the Department of Radiology at Northwell Health in New York, effective AI should be nearly invisible to the user. "If I had to summarize CALM in one sentence, I’d say the best AI may ultimately be the AI that the radiologist doesn't have to think about using," she noted.

In clinical practice, radiologists employ AI tools not only for detection and triage but also throughout the imaging workflow. Goldberg-Stein explained, "We have built tools that synthesize the relevant clinical history for the radiologist. Instead of searching through multiple notes, labs, and prior encounters, the goal is to bring the clinical context to the radiologist at the time of interpretation."

Reducing fragmentation across systems is a major focus, along with ensuring that recommended next steps are implemented effectively. "Increasingly, we're also using AI during the interpretation workflow to suggest next steps in care. Getting patients to the right specialist sooner and preventing them from being lost to follow-up are meaningful outcomes that can make a difference at a larger scale," she emphasized. These recommendations are intended to be patient-centered and should consider the full clinical context, including existing conditions and prognosis.

Paul J. Chang, MD, a professor and vice-chair of radiology informatics in the Department of Radiology at the University of Chicago, compared the interaction between radiologists and CALM AI to driving a car. He stated, "I like driving my car, but I don't like fiddling with the buttons; to me, the optimized human-AI orchestration is that I’m still driving the car, but my AI agent is a helpful passenger."

Chang stressed that effective AI must comprehend what the radiologist is doing and the information needed at any moment. "Without real-time data interoperability and optimized human-machine workflow orchestration, radiologists could become increasingly removed from decision-making," he explained.

Moreover, challenges related to data governance, privacy, and cybersecurity arise as AI becomes more integrated into clinical workflows. He concluded, "AI has to know what I know right now. To work well, I have to also be able to share that orchestration context with the AI agent."

Goldberg-Stein anticipates a future where CALM AI functions as an intelligent layer around the radiologist, efficiently organizing information and supporting critical workflows.