Why AI? The Potential for Computer Minds in Imaging

[vc_column width=”1/2″]Let’s play a quick guessing game. What do the following movies have in common?

I, Robot. A Space Odyssey. The Terminator. The Matrix.

They address the potential threat machines hold to humanity’s way of life.

There is perhaps no greater perpetual source of innovation than the study and development of artificial intelligence. Simultaneously, it has also been a perpetual source of science fiction speculation and an ever-present inducer of job insecurity, albeit in the far, dark corners of our minds.

Many imaging professionals would tell you that there’s no way computers will be able to replace radiologists. We would be inclined to agree with them for now. Take, for example, computer-assisted breast imaging (often referred to as Computer-Aided Detection or CAD)–it is now used on over 90 percent of U.S. mammograms. Early studies suggested that it could help improve detection rates by as much as 20 percent. In fact, many imaging professionals have learned that the value CAD can add is minimal, and now largely ignore the computer results, let alone use them regularly. In fact, one study that appeared in the September 2015 issue of the Journal of the American Medical Association (JAMA) cited concerns that use of CAD actually caused certain radiologists to miss cancer more often than radiologists who had not used CAD.

Although CAD in its current form is far from perfect, the future holds promise for artificial intelligence in clinical decision support. Technology’s past, present, and continued impact on healthcare is undeniable. Since the introduction of the electronic medical record (EMR), the industry has been trending towards increased use of technology. (In all fairness, however, it’s merely a reflection of worldwide technological development. Increased use of technology is in no way unique to healthcare.)

“We’re just barely scratching the surface of using artificial intelligence in the last few years,” said Eliot Siegel, MD, professor and vice chair of research information systems for the University of Maryland Department of Diagnostic Radiology and Nuclear Medicine. “There’s an emergence of increasing interest in the largest companies in the world, including Google, Microsoft, Apple, and IBM, in actually starting to use these technologies for data extraction and evaluation.”

Back to radiology–the computers have a long way to go yet. Since many things such as reporting style and image acquisition are not standardized, the variables for which computers would need to be programmed are virtually endless. Apart from being fitted with the right clinical and peer-reviewed data, any artificial intelligence would need multitudes of images to pull from.[vc_column width=”1/2″]For the foreseeable future, artificial intelligence in medicine can only be a tool. No need to worry about a robot trying to steal your job. The real question we should be asking at this point in time is how can artificial intelligence most effectively be used to complement radiology?

“Prior to 2015, most of what was happening was sort of academic: pilot programs, exploratory, proof of concept-type stuff, and now you’re actually seeing commercial usage,” said Venkat Rajan, global director for Frost & Sullivan’s Visionary Healthcare Program.

He likens a healthcare AI to a digital version of Dr. House.

“At first, [diagnosis] is a complete mystery, it could be one of ten different things. And then he’s able to sort through various issues, you know, illuminate certain factors on why it’s not one of these other conditions, and he’s able to pull something from memory that figures out ultimately what it is, and they can provide the appropriate treatment.”

An effective healthcare AI would be able to sort through a virtually unlimited amount of information and constantly learn from its experiences, remembering each thing it encounters. Additionally, Rajan says that rather than stealing a physician’s job, a robot could help to spare overworked practitioners the fatigue that leads to mistakes.

IBM has shown itself to be at the center of this movement with its Watson Health AI venture. It is being used by industry leaders such as the Cleveland Clinic and Columbia University to process patient data.

It would seem that the future holds great potential for humanity’s work with technology. What are your thoughts on artificial intelligence within healthcare? Agree with us? Disagree? Share your thoughts in the comments below.

The links below are excellent places to start if you wish to learn more about artificial intelligence and healthcare.

Paging Dr. Robot: The Coming AI Healthcare Boom http://www.fastcompany.com/3055256/elasticity/paging-dr-robot-the-coming-ai-health-care-boom

Artificial intelligence in healthcare predicted to grow tenfold in next five years http://www.news-medical.net/news/20160224/Artificial-Intelligence-in-healthcare-predicted-to-grow-tenfold-in-next-five-years.aspx

Posted in Blog, Healthcare IT, Innovation

Comments are closed.