Greg Freiherr, Industry Consultant
Greg Freiherr, Industry Consultant

Greg Freiherr has reported on developments in radiology since 1983. He runs the consulting service, The Freiherr Group.

Blog | Greg Freiherr, Industry Consultant | Artificial Intelligence| September 08, 2017

Must Radiologists Be Prepared To Delegate ... To Smart Machines?

Must Radiologists Be Prepared To Delegate ... To Smart Machines?

Last year, our health system ranked dead last among those of 11 developed nations (including France, Germany, Great Britain and Canada) in calculations done by the Commonwealth Fund, a private foundation that specializes in the analysis of healthcare. This despite seven years of expanded access to healthcare through the Affordable Care Act, also known as “Obamacare.”

No big surprise there. Obamacare — like the debate to repeal and replace it — has been more about insurance than care. To see why simply increasing the number of insured patients won’t make healthcare much better, we have to take a deep dive into the stats.

The U.S. ranks last in healthcare outcomes, according to the Fund. It has the highest rate of mortality “amenable to healthcare,” the Fund stated in a report entitled “Mirror, Mirror 2017: International Comparison Reflects Flaws and Opportunities for Better U.S. Health Care.” And the U.S. has achieved “the smallest reduction in that measure during the past decade,” according to the Fund.

So, clearly, we are not getting the job done. And it isn’t because we’re not spending enough. In 2014, the U.S. spent more than 16 percent of gross domestic product on healthcare, according to the Fund. This compares with between 8 and 12 percent for the 10 other nations in the report. And the gap separating the U.S. from other nations has progressively increased over the years and shows signs of continued widening.

 

Imaging As Catalyst

Rather than debating how to make insurance affordable, we should be looking for ways to make care more effective. A big step in that direction would be the elimination, or at least reduction, of waste. Increased efficiency and making better use of available resources tends to improve performance. Enterprise imaging can do both.

Integrating information sources has the potential to provide diagnosticians and caregivers comprehensive information about patients’ conditions, just as this integration might improve teamwork by those caregivers. Pundits have long promoted such integration.

“We need to move to integrated practice units that encompass all the skills and services,” wrote Harvard business professor Michael E. Porter in the July 9, 2009, issue of the New England Journal of Medicine, a point that Porter reiterated four years later in the Harvard Business Review. Doing so is part of a fundamentally different healthcare strategy to maximize value by delivering the best outcomes at the lowest cost, one that is organized around the needs of patients.

But enterprise imaging will not be enough. It must be part of a culture of change designed and built for efficiency and effectiveness, one that emphasizes meeting the needs of the patient. The technology underpinning this new culture is already bubbling to the surface.

Artificial intelligence (AI) has the potential to help physicians make decisions better and faster. It is already becoming a part of everyday life from smartphones to smart homes.

But before AI can play a serious role in the practice of medicine — and this is the hard part to stomach — some physicians, notably radiologists, may have to be “displaced.”

 

Displacing — Not Replacing

In a JAMA “Viewpoint” published Dec. 13, 2016, an associate professor of radiology at the Hospital of the University of Pennsylvania (HUP) and the chair of innovative medicine at the Scripps Research Institute speculated that some routine tasks performed by radiologists could be automated.

Scripps’ Eric Topol, M.D., and HUP’s Saurabh Jha, MBBS MRCS, MS, urged radiologists to identify “cognitively simple tasks” that smart machines could take over. They cited two such tasks — computed tomography (CT) screening for lung cancer and portable chest radiograms taken in the ICU to confirm the proper positioning of support lines. Delegating simple tasks to machines would allow radiologists to spend more time on challenging cases, where their expertise is most needed. This streamlining of medical practice, if widely applied, holds out the promise for improved efficiency and speedier patient care.

And the use of AI need not be restricted to high-tech imaging. Cameras built into smartphones are already being used to capture everyday events in peoples’ lives.

Might pictures sent via patient portals built into enterprise imaging help in the diagnosis of a skin lesion? Might smart machines review submitted images and select ones that warrant immediate review by physicians, thereby reducing the load placed on diagnosticians and optimizing efficiency?

Similarly, might smartphones be morphed into medical sensors — not only capturing and transmitting vital signs but ultrasound images and video? If so, AI could be leveraged to review these images, spotting cases that deserve careful study by physicians.

There are a lot of “mights” in this. And radiologists will have to go along with it. But the radiology community may soon have little choice. Healthcare costs are mounting. AI might (there’s that word again) make a difference.

It’s important to note, however, that no matter how “artificially smart” medicine becomes, healthcare costs probably will rise. Making medicine more efficient could slow expenditures, but it will probably not decrease them.

The demographics of our population will see to that. Americans are getting older. And older people need more care.

Simply put, enterprise imaging — with or without AI — is not the magic bullet for better or lower cost healthcare any more than fuel-efficient — or electric — cars are the solution to air pollution in a nation with ever more cars. But it can improve efficiency. And healthcare efficiency must be increased.

We can strive to get the most for the money we spend and to deliver the best healthcare in the world.

And those are metrics that really could matter.

 

Related Content on Artificial Intelligence in Radiology

Artificial Intelligence Could Learn From the Medical Imaging Goldmine of the NHS Archives

VIDEO: Machine Learning and the Future of Radiology

How Artificial Intelligence Will Change Medical Imaging

 

Related Content

Machine Learning IDs Markers to Help Predict Alzheimer's

Neurologists use structural and diffusion magnetic resonance imaging (MRI) to identify changes in brain tissue (both gray and white matter) that are characteristic of Alzheimer's disease and other forms of dementia. The MRI images are analyzed using morphometry and tractography techniques, which detect changes in the shape and dimensions of the brain and in the tissue microstructure, respectively. In this example, the images show the normal brain of an elderly patient. Image courtesy of Jiook Cha.

News | Neuro Imaging | September 20, 2018
New research has shown a combination of two different modes of magnetic resonance imaging (MRI), computer-based...
LVivo EF Cardiac Tool Now Available for GE Vscan Extend Handheld Mobile Ultrasound
Technology | Cardiovascular Ultrasound | September 19, 2018
DiA Imaging Analysis Ltd. (DiA), a provider of artificial intelligence (AI)-powered ultrasound analysis tools,...
Exact Imaging Partners to Improve Prostate Cancer Detection With Artificial Intelligence
News | Prostate Cancer | September 19, 2018
Exact Imaging, makers of the ExactVu micro-ultrasound platform, has partnered with U.K.-based Cambridge Consultants to...
SimonMed Deploys ClearRead CT Enterprise Wide
News | Computer-Aided Detection Software | September 17, 2018
September 17, 2018 — National outpatient physician radiology group SimonMed Imaging has selected Riverain Technologie
Acuson Sequoia
News | Ultrasound Imaging | September 12, 2018
Siemens Healthineers announced the first global installation of its newest ultrasound system, the...
Veye Chest version 2
News | Lung Cancer | September 11, 2018
Aidence, an Amsterdam-based medical AI company, announced that Veye Chest version 2, a class IIa medical device, has
Sponsored Content | Case Study | Information Technology | September 07, 2018
Established in 1970, Sovah Health – Martinsville, Va., resides in the foothills of the beautiful Blue Ridge Mountains...
Feature | Population Health | September 07, 2018 | By Jeff Zagoudis
Over the last several years in the U.S., healthcare providers have been trying to shift their focus to more preventive...
Sponsored Content | Case Study | Information Technology | September 07, 2018
One of the Northeast’s major teaching hospitals is an international leader in virtually every area of medicine. It has...
The CT scanner might not come with protocols that are adequate for each hospital situation, so at Phoenix Children’s Hospital they designed their own protocols, said Dianna Bardo, M.D., director of body MR and co-director of the 3D Innovation Lab at Phoenix Children’s.

The CT scanner might not come with protocols that are adequate for each hospital situation, so at Phoenix Children’s Hospital they designed their own protocols, said Dianna Bardo, M.D., director of body MR and co-director of the 3D Innovation Lab at Phoenix Children’s.

Sponsored Content | Case Study | Radiation Dose Management | September 07, 2018
Radiation dose management is central to child patient safety. Medical imaging plays an increasing role in the accurate...