More on Analytics

With regulatory barriers falling, AI will see more clinical adoption in 2019

AI is on the verge of being de-risked, so the market is poised to flourish as caregivers look to ramp up their patient care efforts.

Jeff Lagasse, Associate Editor

Technology sometimes creeps its way into healthcare in fits and starts, and this may well be the case with artificial intelligence and machine learning, which are poised to transform the industry but haven't quite reached the level of outright pervasiveness.

In the view of Yann Fleureau, co-founder and CEO of Cardiologs, 2017 and 2018 were the years of regulatory validation for AI, with companies bursting through the regulatory barrier with diagnostics programs and other applications that proved their effectiveness.

Now looking toward the future, he expects that 2019 will be the year of commercial and clinical validation, meaning the solutions that emerge can be effectively used to manage patients.


Validation is moderately related to risk, said Fleureau. AI is on the verge of being de-risked, which is basically a code-word for breaking down those regulatory barriers -- so the market is poised to flourish, allowing caregivers and healthcare providers to ramp up their patient care efforts.

"There's a broader picture," said Fleureau. "At the end of the day, caregivers do not care about AI. They care about patient safety. AI and technology enables the benefits. It's no different from another non-AI based technology.

"From our experience, we have not experienced customer objection specific to AI," he said. "As long as you come with a value proposition that promises to bring benefits to the users, in terms of cost and clinical efficiency -- to me, that's really the point."


One of the reasons the AI market is primed to take off in healthcare is that it holds the potential to automate mundane reporting requirements, which have wrested clinicians from patient care and forced them to spend more of their working hours reviewing information that could be better left to machine learning.

"Caregivers spend most of their time looking at false alarms," said Fleureau. "That's really what's mundane. The tedious part of their job is to review data, which does not benefit patients.

"By bringing a deep learning-enabled technology which is radically superior, you demonstrate you have the same level of safety -- no false negatives -- while reducing the administrative burden. You save caregivers' time so they can focus on the true items that require medical attention."

Both AI and machine learning are in a prime position to alter clinical workflows and physician training. And with the market growing the way it is, implementation is inevitable. A recent Accenture report estimated that the AI health market will hit $6.6 billion by 2021. That's up from $600 million in 2014.

The good news is that AI will likely not come for providers' jobs -- at least for now.

In the analysis, the author notes that innovations in various industrial revolutions have always created new jobs even as they've taken old jobs away. What makes the AI revolution different is that it has the potential to affect white-collar jobs.

Initially AI in healthcare would primarily affect office workers, such as those in data processing. Though more highly trained professionals could also be affected, the switch so far seems to be happening in a way that shows AI to be a tool more so than a threat, as professionals can now learn how to benefit from its powerful predictive powers.

In some cases, the technology could be used to help fill the physician shortage that is even now gripping many parts of the country, and is expected to get worse.


Workflow improvement and drug discovery are two use cases for AI that will become even more evident as AI in healthcare advances, said Fleureau. When it comes to clinical support, the areas most primed to receive the technology will likely be imaging and diagnostics.

Deep learning technology was developed largely for those use cases, so most of the companies that have broken through the regulatory barrier so far have been imaging- and diagnostic-centric. They'll be the first areas to lead the way to large-scale adoption.

Teaching hospitals will likely be the first to use the technology in a pervasive way, said Fleureau, because they're typically based in urban areas and have more patient flow.

"From my perspective they will use it first, because it's both in the interest of the innovators, and it's the traditional way to bring products to market," said Fleureau.

"Caregivers and AI speak the same language, which is the language of statistics," he said. Pretty much anywhere there is data, there will be AI."

Twitter: @JELagasse

Email the writer: