Features

What Are the Benefits of Natural Language Processing Technology?  

Using natural language processing technology, researchers can sort through unstructured data to improve patient care, research efforts, and disease diagnosis.  

Source: Getty Images

- To deliver quality care and positive patient outcomes, researchers and clinicians need comprehensive patient data and medical literature. 

 However, since 80 percent of essential data lies in unstructured clinical notes, scientific papers, and conference articles, gaining immediate access to clinical information is difficult. 

Researchers can utilize artificial intelligence methods to sort through unstructured data by using natural language processing. The data can then provide valuable insights into patient care, research efforts, and disease diagnosis.  

What is natural language processing? 

Natural language processing is the overarching term that describes using computer algorithms to identify key components in everyday language, exact meaning from the unstructured or written input, and turn it into usable data. NLP requires the use of artificial intelligence, computational linguistics, and other machine learning methods. 

 Some of the specific tasks for NLP systems can include: 

  • Summarizing lengthy blocks of narrative text, including clinical notes or academic journal articles, by identifying key concepts or phrases present in the material 
  • Mapping data elements present in unstructured text to structured fields in an electronic health record to improve clinical data integrity 
  • Converting data in the other direction from machine-readable formats into natural language for reporting and educational purposes 
  • Answering unique free-text queries that require the synthesis of multiple data sources 
  • Engaging in optical character recognition to turn images, like PDF documents or scans of care summaries and imaging reports, into text files that can then be parsed and analyzed 
  • Conducting speech recognition to allow users to dictate clinical notes or other information that can then be turned into text 

Many NLP systems “learn” over time, reabsorbing results from previous usages as feedback regarding which results were accurate and which did not meet the research needs. 

How can NLP improve patient care and research? 

To better manage unstructured data and research efforts, Georgetown University Medical Center adopted artificial intelligence-based text-mining tools in electronic health records. The tool allows physicians to quickly search through large amounts of medical literature to support real-time clinical decision-making.  

“I work with physicians a lot and talk to them about data searching. Their main complaint is that they can’t pull out the precise information they want quickly when they do a search. The search is very imprecise. It takes them too long and too much work to get to the information that they want,”” Clinical Informaticist at Georgetown university, Johnathan Hartmann, told HealthITAnalytics.  

The Linguamatics text-mining tool uses natural language processing to sort through text for specific key phrases. The pulled information can then identify the best course of treatment for patients.  

Using that same AI technology, research teams can search literature and medical records to discover genes associated with certain diseases to improve their understanding of molecular processes and to advance drug targeting.  

While traditional text-mining tools perform similar tasks, Linguamatics significantly lightens the load on physicians.  

“The tools allow clinicians to search the text of a larger number of articles very quickly and efficiently, pulling out the information that they’re interested in precisely. Traditional search engines like PubMed, might search and pull up 50 articles, some of which may contain the information they’re looking for,” Hartmann continued.  

“However, clinicians would have to read through either the abstract or even the full text themselves to identify maybe one or two articles that have the information they’re looking for. Using Linguamatics, they could identify those two articles immediately and not have to sift through 50 articles.” 

Additionally, Linguamatics uses natural language processing capabilities to search the entire text of an article to identify concepts and relationships in literature to deliver high-quality care. Traditional text-mining search methods without natural language processing capabilities are unable to perform the same tasks. 

Using comprehensive data allows physicians to provide high standards of patient care. According to Hartmann, there is plenty of information in unstructured data and literature, which can play an important role in clinical decision-making.  

How can NLP assist with disease diagnosis? 

In addition to identifying the correct treatment options for patients, NLP can also assist clinicians in disease diagnosis.  

 A recent study by Kaiser Permanente demonstrated the value of natural language processing (NLP) technology with clinicians identifying more than 50,000 patients with aortic stenosis. 

The study was conducted by Matthew Solomon, MD, PhD, a cardiologist at The Permanente Medical Group and a physician researcher at the Kaiser Permanente Division of Research in Oakland, California. 

According to Solomon, while healthcare is currently in an era of big data and data analytics, it remains challenging to identify patients with complex conditions such as valvular heart disease, creating challenges with it comes to studying the disease, tracking practice patterns, and managing population health

“Currently, health systems track patients using diagnosis or procedure codes, which are mostly created for billing purposes. These can be very non-specific and are not useful for clinical care or research,” Solomon told HealthITAnalytics.  

“Without accurate and systematic case identification, population management and research on valvular heart conditions and many other complex conditions aren’t possible. We set out to tackle this problem by developing natural language processing algorithms that make it possible to teach a computer how to do this for us.” 

The research team trained the NLP to sort through over a million electronic medical records (EMR) and echocardiogram reports to detect certain abbreviations, words, and phrases associated with aortic stenosis. 

Within minutes, the software recognized almost 54,000 patients with the conditions, a process that would have likely taken years for physicians to perform manually. 

“It was a magical moment when we were able to apply our developed and validated algorithms on our entire population and to then identify our large cohort of patients with aortic stenosis,” Solomon said. 

“We could immediately imagine a not-too-distant future where these methods could be used to take population management, which Kaiser Permanente Northern California has excelled at for the past two decades, to the next level.” 

As artificial intelligence continues to grow in healthcare, Solomon said providers should be confident that investing in new technologies will significantly improve patient outcomes

“These AI techniques will be able to assist doctors and other providers to care for their patients in ways that were not previously possible,” Solomon said. 

With NLP technology, researchers can enhance patient care, research efforts, and disease diagnosis methods.