How Can Healthcare Leverage Natural Language Processing for Medical Records?

Earlier this year, Microsoft announced the $19.7 billion acquisition of cloud and AI software leader Nuance, marking the tech giant’s increased expansion into healthcare. Nuance offers AI tools that integrate with EHRs to support data collection and clinical note composition.

“Nuance provides the AI layer at the healthcare point of delivery and is a pioneer in the real-world application of enterprise AI,” said Microsoft CEO Satya Nadella in an April statement marking the deal. “AI is technology’s most important priority, and healthcare is its most urgent application.”

As healthcare systems continue to find ways to solve clinician burnout from EHRs, AI-based solutions can offer a path to ease staff frustration and increase time for patient care.

DISCOVER: How can AI increase efficiency in healthcare, even in a pandemic?

The Rise of NLP in Healthcare

Natural language processing was largely an academic topic before it took off in the commercial space due to a massive increase in computing power, the deployment of cloud computing, and heavy investment from tech companies into developing algorithms that can derive meaning from human speech, according to Mahi Rayasam, a partner in McKinsey’s Detroit office and co-leader of Healthcare Analytics by McKinsey.

“Ten years ago, if I had — with my non-American accent — spoken to the precursor to Siri, it would not have understood me. But now, it understands better,” he says.

Clinical data, on the other hand, is very complex, Rayasam says. From a semantic process standpoint, consumer data is relatively easy to understand, such as asking Siri about the local forecast or telling Alexa to order more detergent online. But clinical data — including physicians’ notes about a visit, emails and specialist images — often contain critical information, the meaning of which can be harder to capture through an automated process.  

“In healthcare, to build a training set is extremely hard,” Rayasam says. “Creating the training set requires that you have people with deep clinical knowledge who are going through rates of data and tagging them appropriately, and then creating data sets that the NLP can learn on. That’s a big difference when you compare clinically oriented NLP or clinically oriented bot versus a consumer-oriented bot.”

Interest in NLP applications in the healthcare space has grown, driven by the move to EHRs and the improvement in accuracy levels for increased entity extraction and document specification, Talby adds. Though clinical NLP use is showing promise in the industry, there are still years to go before widespread relief of clinician burnout with EHRs.

“There’s a very big difference between having a good result on an academic paper and having something that actually works in many hospitals,” Talby says.

Still, NLP applications have gained traction in healthcare in the past five years, Rayasam says, starting with medical transcription services, inputting clinician notes into an EHR and making sense of those notes.

Such applications include Nuance’s Dragon Medical One and Dragon Medical Practice edition. Concord Hospital in New Hampshire used Dragon technology when it moved to Cerner’s Millennium EHR system, enabling clinicians to provide dictation from any workstation or smartphone, which allowed the health system to save money by dropping the use of phone-based transcription services, HealthTech reported last year.

READ MORE: Language processing tools improve care delivery for providers.

Minneapolis-based Allina Health also turned to NLP technologies, creating significant cost savings for the health system and an expedited workflow for providers.

“I can say, ‘go to the most recent labs,’ and the computer will navigate there for me,” Dr. David Ingham, chief health information officer at Allina, told HealthTech. “I can say, ‘Order a basic metabolic panel,’ and it will tee that up.”

What Could Future NLP Use Cases Bring?

The increased adoption of NLP applications in healthcare is on the industry’s radar. In an online survey with results released earlier this year, John Snow Labs found that 36 percent of respondents who are healthcare and life sciences practitioners plan to have NLP technologies in place by the end of 2021.

One potential use case — combining NLP with computer vision — would allow for techniques of multimodal learning, combining what a computer can see, read and hear, Talby said.

Other future use cases include “bringing more intelligence into the search” process within a single patient’s EHR, thus helping with better care recommendations, Rayasam adds. NLP applications could also be used to simplify administrative processes such as prior authorization.

Because healthcare systems hold massive amounts of data, the combination of NLP with other AI capabilities can offer a world of solutions that could better support clinical decision-making and help physicians better focus on their patients instead of their device screens.

MORE FROM HEALTHTECH: Why nurses are essential to AI integration in healthcare.

“NLP is not something that can make an impact by itself, but when combined with other cutting edge computing techniques it can simplify a lot of processes, discover underlying patterns or underlying conditions, and hasten decision-making that is appropriate for the right outcomes,” Rayasam says.

For healthcare organizations looking to deploy an NLP application, Talby says, “as with all technology, don’t use this because it’s the next, new shiny object.”

The interest in NLP in healthcare is growing, he adds, with potential use cases in the pharmaceutical sector as well, such as automation to match patients with clinical trials.

“These technologies are very exciting and could potentially change the future of how we manage healthcare in the nation,” Rayasam says.