Skip to main content

How Drexel University is using Natural Language Processing to improve clinical healthcare

Linguamatics NLP platform supports medical research and patient care delivery

Natural Language Processing (NLP) is used to transform text and unstructured data into valuable, real-life, outcomes. Generally in healthcare NLP is still in a relatively early stage of adoption. However, some organizations are moving forward towards full success in using NLP to deliver enhanced healthcare research and clinical processes.

Walter Niemczura, the director of application development at Drexel University College of Medicine in Philadelphia, is one of the individuals driving the ongoing initiative to improve healthcare research. Niemczura began working with Linguamatics seven years ago, in order to identify patients with certain characteristics that were well represented in unstructured clinical notes from Electronic Health Records (EHRs). Niemczura realized that the discrete data they had been working with wasn’t going to be enough to really advance and support research and patient care efforts.

"Linguamatics NLP was a huge time-saver. When you’re looking at hundreds of thousands or millions of patient records, the value might be not the ones you have to look at, but the ones you don’t have to look at." Walter Niemczura, director of application development, Drexel University College of Medicine

NLP allows for faster and better EHR analysis

As Niemczura explained in a recent interview with Healthcare Informatics, if you are looking at thousands of patient charts, you might miss out on crucial information presented in the EHR clinical note records. While anybody in an IT department can identify a diagnosis, they might miss out on information from EHRs that are contained in non-discrete fields, such as patient notes. Different providers may document the same diagnoses in many different ways - this process is seldom standardized. Linguamatics NLP technology can provide an augmented intelligence (AI) solution, which narrows down the number of patients’ charts clinicians need to look at.

Using Linguamatics, instead of residents having to review at 5,700 charts, they were able to reduce manual review to only 1,150 (1 in 6 charts) - a manual chart review reduction of over 80%, a huge saving of time and related costs for the Drexel team. A complete analysis of the information within the EHR also allows for far more accurate decisions.

Linguamatics also provided on-site training to the Drexel University team, reducing the learning curve required to introduce the new technology. Additionally, as Niemczura stated in his interview, by using Linguamatics NLP, Drexel is improving accuracy in a far more efficient way - looking at patient trials on the research slide; and on the clinical side- delivering better quality of care and updating billing, based on more accurate data management.

"We could do word search in Microsoft Word; but the word “erosion” by itself might not help. You have to structure your query to be more accurate, and exclude certain appearances of words. And Linguamatics is very good at that." Walter Niemczura, director of application development, Drexel University College of Medicine.

DOWNLOAD CASE STUDY

Read the full interview by Mark Hagland, Editor-in-Chief of Healthcare Informatics here or contact us to discuss your medical research and patient delivery care initiatives, and learn more about the Linguamatics NLP platform.

Ready to get started?

Request a Demo

Questions? Ask our experts