Top takeaways from RSNA 2019: The human side of imaging

Clean Sweep Live Auction on Wed. Jan. 22nd. Click to view the full inventory

Top takeaways from RSNA 2019: The human side of imaging

December 11, 2019
Business Affairs CT MRI X-Ray
RSNA 2019
By the HCB News editorial team

Last week the radiology world converged on McCormick Place in Chicago for the annual meeting of the Radiological Society of North America (RSNA). As one might have expected, AI was a dominant topic. There was even an entire exhibit hall dedicated to it.

And yet, in many ways, the human aspect of radiology was even more prominent.

It felt almost as if, after all these years of asking rhetorical questions about computers replacing humans, the industry is finally getting comfortable with the concept of AI and can consider it in nuanced, practical terms. This, the editorial team at HealthCare Business News agrees, is good news for everyone.

The best diagnostic nuclear imaging systems, parts and service available

Universal Medical provides the very best new & refurbished gamma cameras, quality parts & repair services. We also rebuild & replace camera detectors, move camera systems across town or across the country. Call us at 888-239-3510


AI marketplaces and workflow integration
From Philips' work-in-progress IntelliSpace AI Workflow Suite to IBM's Imaging AI Marketplace, it felt like every big company debuted an AI "App Store" or platform to integrate curated AI algorithms into the radiology workflow.

Whereas last year's show might have featured the release of an AI solution for flagging critical diagnoses, this year the talk was all about how to make the use of machine learning seamless for radiologists.

“The key to the game is going to be really translating that directly into the clinical workflow, finding ways to be able to really put that in, so that it’s not just something that’s adjacent or added, so customers don’t have to contract with a million different vendors,” Kevin Lev, the AI solutions marketing lead at Philips, told HCB News during an interview.

Overcoming bias and encouraging diversity
Bias in radiology is no longer just a challenge in the workplace, it's also being exposed in the data sets that are intended to help automate decision-making.

In a session on the ethical challenges of AI, Judy Gichoya, assistant professor of IR Informatics at Emory University School of medicine, spoke about the recent news that racial bias was embedded in a widely-used commercial risk algorithm used to manage population health.

There were also multiple sessions about bias in recruitment and hiring of female radiologists and physicians of color.

During one session on diversity and inclusion in radiology, it was noted that about 50 percent of medical school graduates are women, but only 27 percent become radiology residents, and only 15 percent of medical school graduates who identify as underrepresented minorities pursue careers in radiation oncology and diagnostic radiology.

You Must Be Logged In To Post A Comment