By Aine Cryts

Ryan Lee, MD, section chief of neuroradiology at Philadelphia’s Einstein Healthcare Network, remembers the first time he heard about the role of a radiologist. The father of three says he first got interested in radiology during his college years when he was working with his family doctor who had an x-ray machine.

“He would hire a radiologist to come in once a week to read [x-rays] and always grumbled about having gone into the wrong field. Reading the x-rays was always my favorite part of the day,” remembers Lee, who also serves as vice chair of quality and safety at Einstein Healthcare Network.

Historically, the radiologist’s role was primarily to interpret radiology exams, he says. But much has changed. “While obviously important, this is but one role of [today’s] radiologist,” adds Lee, who expects additional changes in the role of the radiologist in the future.

AXIS Imaging News recently discussed the role of the radiologist in healthcare today. What follows is a lightly edited version of that conversation.

AXIS Imaging News: What does the modern radiologist’s role include?

Ryan Lee: The modern radiologist must play a central role in all aspects of imaging, beginning with the initial decision from a referring physician to order a study. This includes assisting the referring physician in selecting the most appropriate study using clinical decision support; determining the most appropriate protocol and parameters for a given patient; and assisting the referring physician with the most appropriate next step in the management of the patient, given the results of the examination.

In addition, a radiologist’s more direct interaction with the patient—and maximizing patient satisfaction and safety while ensuring quality examinations—is paramount. This holistic approach to radiology will become increasingly important as healthcare continues to migrate more toward value-based care and away from volume-based care.

AXIS: Explain how you view valued-based care as a radiologist.

Lee: In general, we think of value-based care as value=quality/cost. Or put another way, value-based care is the best possible quality care using the least amount of resources. But it’s not as simple as that. It used to be that “doctor knows best,” but we now realize that patient-centered care must be approached more holistically from many viewpoints and data points.

AXIS: How is access to information an important part of value-based care?

Lee: Healthcare is becoming more inclusive in terms of the types of information we consider when diagnosing what’s wrong with a patient, but information also helps guide the right course of treatment.

Patients may have special circumstances unknown to the doctor, and acknowledging these factors is part of the goal of value-based care. At the same time, with value-based care, the business of healthcare is changing via new reimbursement models. Fee-for-service is going away.

Quality is more important than quantity, but quality is still in the beginning stages in healthcare. Reducing variability in consistently reproducible ways hasn’t really been addressed. That translates to a real opportunity for radiology. Many healthcare networks are now incentivized to exclude specialists like radiology, which means radiology needs to be more vocal about imaging’s role in value-based care and the key metrics, such as radiation dose, that we can improve upon.

AXIS: What’s one of the most profound changes you’re seeing in radiology today and what impact will it have on the specialty?

Lee: Without question, artificial intelligence and, more specifically, machine learning will have a profound change in the field of radiology. Artificial intelligence and machine learning have the potential to change fundamentally how radiologists practice radiology.

From a diagnostic perspective, artificial intelligence and machine learning can be of tremendous aid to the radiologist in identifying and even classifying lesions. Other benefits of using artificial intelligence and machine learning include non-diagnostic applications, such as scheduling, protocoling, and workflow optimization. Ultimately, artificial intelligence and machine learning will increase the productivity of radiologists, allowing them to do more than was previously possible.