How Can Healthcare Leverage Natural Language Processing for Medical Records?

As automated solutions continue to gain traction in the industry, here’s how a specialized branch of artificial intelligence is showing promise in improving EHR usability. The healthcare industry continues to search for and deploy feasible solutions for physician burnout linked to electronic medical and health records. According to the American Medical Association, physicians can spend up to two hours in an EHR system for every hour they spend with their patients.

Though scribes and medical assistants can be helpful in easing some burdens linked to the data entry process, there are still many issues, including a focus on documentation geared toward billing rather than patient care, information overload and difficulty navigating a system quickly.

“Part of the issue with documentation is that we are asking doctors to document things in a very specific and unnatural way,” says David Talby, CTO of healthcare artificial intelligence company John Snow Labs. There are quality metrics to consider, and EHR documentation is often written the way it is for insurance purposes, Talby adds.

But the text-rich nature of an EHR system means that it can be well suited for an automated process such as natural language processing, a specialized branch of AI that allows computers to understand unstructured written or spoken data. And NLP’s promise to improve medical record usability has spurred a lot of business interest in the healthcare industry.