Huntsman Cancer Institute (HCI) at the University of Utah is a nationally recognized cancer center that relies heavily on data for its research studies. Because a vast amount of critical patient information is stored in unstructured formats such as clinical notes and pathology reports, finding specific data is often challenging, to say the least—not to mention costly and time consuming.

For years, HCI had compiled information manually or with rudimentary natural language processing (NLP) tools, but surely there was a better way?

After encountering Linguamatics at an informatics conference and learning more about its NLP tools and Linguamatics text mining solution, the HCI research informatics team realized that this is what it had been searching for. To test the system, HCI used Linguamatics NLP platform in a project on breast cancer; it found that data capture was much, much faster, and using NLP improved access to higher quality data.

Since that initial success, HCI has expanded its use of the platform and developed NLP tools for multiple other conditions. It can now provide investigators with the quality data they need when applying for grants, writing papers, or identifying cohorts for specific studies. HCI has also been able to share data and collaborate with other institutions, to advance research and enhance disease understanding, and ultimately achieve better patient outcomes.

“Linguamatics NLP platform is the driver for collecting high quality data and making the process more efficient.” Samir Courdy, Director of Research Informatics Shared Resource
and Chief Research Information Officer, HCI


We don’t know what we don’t know. 

“The three great essentials to achieve anything worthwhile are, first, hard work; second, stick-to-itiveness; third, common sense.”

– Thomas A. Edison

Recently I had the honor to spend some time with some amazing people that are committed to quality in healthcare at the at 2018 HQI Conference in Huntington Beach- yes the sacrifices I must make- what an amazing venue! There, I learned that most of the attendants have been doing their reporting for quality via long hours of manual effort. I look at this in two ways…

1) I applaud such valiant efforts to achieve something worthwhile; and

2) are they unaware that technology can decrease their burden? These hard working professionals certainly cover the hard work and stick-to-itiveness. As for the third essential, this isn’t a case of a lack of common sense, it’s simply a case of…”we don’t know what we don’t know”.  

How do you ensure healthcare quality when the priority is quantity?

Data is everywhere. When it comes to data, what is one person’s garbage is another person’s treasure. It simply has to do with what question you have to ask. And when it comes to quality and reporting metrics it’s best to listen to the experts.  The National Committee for Quality Assurance (NCQA) exists to improve the quality of health care...they even say so on their web page.


Pondering DNA at The Eagle, Cambridge

I was recently privileged to have a pint of Guinness at the Eagle in Cambridge with some colleagues after work in our U.K. office. The Eagle is a historically significant pub in the area of DNA. Two things came to mind:  1) What is it with the number 51 and controversies? Area 51 and Photo 51 both bring up their own issues...and if we combine the two it would get really interesting...Alien DNA. Now that would be a good pub conversation! Especially here at the Eagle, where James Watson and Francis Crick theorized DNA’s helical structure.  And 2) I wish Rosalind Franklin could have lived to see how things are evolving in precision medicine.


Linguamatics is pleased to congratulate US healthcare system Mercy on their recent award win at the 12th Gateway to Innovation conference. Mercy won the Innovative IT Project of the Year Award for using Linguamatics I2E Natural Language Processing (NLP) solution to extract clinical analytics insights from their Electronic Health Records (EHR) notes for cardiac patients.

Mercy Technical Services provides contract research services for medical device and pharmaceutical clients to support use of real world evidence (RWE) in Food and Drug Administration submissions. This award recognizes a project that demonstrates value or impact to the organization by solving a business problem or by addressing a specific strategic objective for the company.

NLP used to Extract Real World Evidence from EHRs

As a large health system with a mature and consolidated Epic EHR system, Mercy has a significant data set of patient treatments and outcomes. There is a multitude of information documented in the EHR, such as lists of specific symptoms, diagnoses derived from echocardiogram reports, and certain benchmarking classifications. Since typically 80% of this information is unstructured text, many valuable clinical insights are unavailable in discrete fields, and therefore vital patient information can be trapped when making clinical decisions.

NLP text mining platforms like Linguamatics I2E extract information from unstructured text-based EHRs and transform it into actionable insights that can be placed into a dataset and analyzed.


The Philosophy of Yin and Yang

The idea itself is relatively simple- all things exist in a state of entanglement and contradiction to themselves. Like my favorite geeked out analogy - Star Wars with the light (Yin) and the dark side (Yang). If you recall- Luke (Yin) and Darth Vader (Yang) both always had one foot over the line that delineates the dark side from that of the light. Either character could have changed their sides but ultimately there must always be a balance in the Force.

Yin, Yang and Star Wars - What Does That Have to Do With Modern Healthcare?!

It’s about balance. Healthcare organizations are businesses. A business by its very nature needs to make a profit to continue its existence. Healthcare organizations just happen to be in the business of caring for people. It is a business that by all ethical principles should not be performing unnecessary procedures to stay out of the red and yet this is still done by some organizations.