4 New Categories for Faster Image Processing

January 6, 2021
Whitney J. Palmer

As imaging volumes continue to rise, experts offer a targeted solution that could lead to faster throughput.

The concept of creating targeted therapies for a patient’s individual needs is not new – but what about doing the same thing with imaging services and a radiologist’s expertise?

It might go against the grain of what providers usually strive to do. Most want to provide the highest level of diagnostic evaluation and the most detailed reporting possible. But, doing so does not necessarily offer the best value to the patient, and as imaging volumes hold steady or increase, it could create problems for the radiologist.

“If current volume trends continue and if there are no transformative technological leaps in how radiologists interpret studies, the specialty will be increasingly challenged,” wrote a team led by Nadja Kadom, M.D., associate professor of radiology and imaging sciences at Emory University School of Medicine.

In an article published in the January Journal of the American College of Radiology, an issue dedicated to focusing on and tackling provocative, uncomfortable, and debatable topics, Kadom’s team posited pivoting to a work model that more directly matches imaging services to the clinical context. It would be a way to provide better service with less resource and time investment.

In the next 30 years, they said, radiology will face an uphill climb. By 2050, the American population is expected to top 438 million people, and far more quickly – by 2032 – experts anticipate the country will see a physician shortage of 122,000, including 39,100 specialists. That dearth includes radiologists. And, with greater employment of nurse practitioners and physician assistant increasing imaging orders by 1.3 times, meeting patient needs in the most timely, effective way will be a tall order.

There could be way to side-step this problem, however, said Richard Duszak, M.D., Jr., one of the study authors and vice chair for health policy and practice in Emory’s radiology and imaging sciences department, and it results in providers saving time and energy. It is an aspiration that can potentially re-invent radiology over time through discussion and collaboration.

“Radiologists – and the entire healthcare enterprise more broadly – need to begin having thoughtful and brutally realistic conversations about how we can all collectively meet the needs of an expanding population in an equitable manner as dollar-constrained healthcare resources become increasingly constrained,” he said. “What we have proposed is a clinical-needs-based (rather than one-size-fits-all) approach to providing imaging services in a manner that would most effectively utilize radiologists’ time and expertise.”

This approach, the team laid out, would open the door to more appropriate techniques, protocols, and levels of reporting by separating imaging studies into categories that differ based on level of technology use, level of reporting, and clinical intent. This move, which is intended to be provocative, would create four new categories: emergency, surveillance, precautionary, and exclusionary imaging studies.

For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.

Emergency: Faced with the constant time crunch to identify the reason behind a patient’s symptoms, emergency radiologists looking for all pertinent findings are consistently in jeopardy for medicolegal action if they miss a finding or render a misdiagnosis.

Under this targeted model, the team said, emergency radiologists could limit themselves to the specific questions related to the patient’s acute need, focusing on relevant findings and potentially producing better throughput.

“In light of such circumstances, perhaps radiology practices should prioritize addressing specific clinical questions, rather than seeking exhaustive documentation and reporting of the variety of incidental findings emergency radiologist encounter,” they said. “Such an approach would establish more realistic expectations for radiologists working under significant time constraints.”

Surveillance: In a similar way, the team said, surveillance imaging can be circumscribed around findings associated with a condition that radiologists already anticipate detecting. Providers would be able to focus their efforts on areas closest to the affected regions of the patient’s body.

“Surveillance imaging protocols and reporting templates could be tailored to detect and report very distinct findings, thus permitting the interpreting radiologist to focus on the most likely range of findings and imaging outcomes, rather than a full and broad range of diagnostic possibilities unrelated to the surveilled condition,” they said.

For example, in this situation, a radiologist could use an abbreviated MRI protocol without contrast to follow cystic pancreatic lesions.

Precautionary: This type of imaging is not typically considered standard-of-care, but the need for it can arise, most likely, in emergency settings. The studies can be conducted to settle patient anxieties or to satisfy a need for defensive medicine and can reasonably be accomplished with fewer imaging or faster protocols, the team said. A checklist approach could help providers safely exclude various diagnoses.

“In such precautionary examples, expectations for diagnostic performance may diverge from those which one would expect with full diagnostic studies, also allowing precautionary studies to be done with higher throughput than full diagnostic studies,” they said.

Exclusionary: Even this infrequent type of imaging can be pared down, the team said, allowing the radiologist to concentrate on very pointed questions.

“Conceivably, imaging protocols could be tailored and shortened to answer very specific questions, depending on the particular diagnoses that must be excluded for a given presentation,” they explained.

Implementing this type of imaging categorization holds several benefits for radiologists. Referring physicians would be able to accelerate the process by specifically ordering a surveillance or precautionary study, largely alleviating the medicolegal risk radiologists take when using targeted or abbreviated protocols. In fact, the team said, such a system could also encourage greater communication between radiologists and referring providers to identify the clinical scenarios best suited for this targeted approach.

But, shifting to a more stream-lined, targeted method of imaging will not necessarily be easy, the team admitted. Many radiologists may be reticent to scale back on imaging because doing so might inhibit their ability to catch potentially significant pathologies in earlier stages when treatment efforts can be more successful. In addition, categorizing imaging into these buckets could also lead to drops in reimbursement.

In fact, Duszak said, successful implementation would require a payment system overhaul, moving from traditional fee-for-service and one-size-fits-all billing requirements which could potentially necessitate more work than is clinically necessary, as well as tort reform. Consequently, it is imperative that all stakeholders be involved in the decision-making process to balance the competing goals of increased patient access and reporting.

Failing to change the industry’s approach to imaging could put providers in a position where they face an ever-growing mountain of scans, he said.

“As the demand for imaging increases in a world in which imaging is more complex and patients are sicker, radiology practices are increasingly struggling to meet patient service needs,” he said. “Blanket ‘just say no’ responses to innovation, such as the judicious supervised use of artificial intelligence and non-physician provider, continues us on our current trajectory – which many consider unsustainable.”

Ultimately, he said, continuing the course could make the specialty less attractive to medical students, effectively choking off the talent pipeline.

Even with those challenges and the imperfections of the proposed categorization system, the team said, they hope it will open the door to finding new ways to process the industry’s workload.

“Shifting the imaging paradigm by defining radiology testing categories based on clinical intent and imaging and reporting detail can open new avenues for managing high volumes,” they said. “Although the paradigm shift proposed here may not be the ultimate remedy, we hope that it can invite discussion, alleviate radiologist burnout, and ultimately improve patient care.”