Drexel University Moves Forward on Leveraging NLP to Improve Clinical and Research Processes | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Drexel University Moves Forward on Leveraging NLP to Improve Clinical and Research Processes

January 8, 2019
by Mark Hagland, Editor-in-Chief
| Reprints
At Drexel University, Walter Niemczura is helping to lead an ongoing initiative to improve research processes and clinical outcomes through the leveraging of NLP technology

Increasingly, the leaders of patient care organizations are using natural language processing (NLP) technologies to leverage unstructured data, in order to improve patient outcomes and reduce costs. Healthcare IT and clinician leaders are still relatively early in the long journey towards full and robust success in this area; but they are moving forward in healthcare organizations nationwide.

One area in which learnings are accelerating is in medical research—both basic and applied. Numerous medical colleges are moving forward in this area, with strong results. Drexel University in Philadelphia is among that group. There, Walter Niemczura, director of application development, has been helping to lead an initiative that is supporting research and patient care efforts, at the Drexel University College of Medicine, one of the nation’s oldest medical colleges (it was founded in 1848), and across the university. Niemczura and his colleagues have been partnering with the Cambridge, England-based Linguamatics, in order to engage in text mining that can support improved research and patient care delivery.

Recently, Niemczura spoke with Healthcare Informatics Editor-in-Chief Mark Hagland, regarding his team’s current efforts and activities in that area. Below are excerpts from that interview.

Is your initiative moving forward primarily on the clinical side or the research side, at your organization?

We’re making advances that are being utilized across the organization. The College of Medicine used to be a wholly owned subsidiary of Drexel University. About four years ago, we merged with the university, and two years ago we lost our CIO to the College of Medicine. And now the IT group reports to the CIO of the whole university. I had started here 12 years ago, in the College of Medicine.

Webinar

Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

And some of the applications of this technology are clinical and some are non-clinical, correct?

Yes, that’s correct. Our data repository is used for clinical and non-clinical research. Clinical: College of Medicine, College of Nursing, School of Public Health. And we’re working with the School of Biomedical Engineering. And college of Arts and Sciences, mostly with the Psychology Department. But we’re using Linguamatics only on the clinical side, with our ambulatory care practices.

Overall, what are you doing?

If you look at our EHR [electronic health record], there are discrete fields that might have diagnosis codes, procedure codes and the like. Let’s break apart from of that. Let’s say our HIV Clinic—they might put down HIV as a diagnosis, but in the notes, might mention hepatitis B, but they’re not putting that down as a co-diagnosis; it’s up to the provider how they document. So here’s a good example: HIV and hepatitis C have frequent comorbidity. So our organization asked a group of residents to go in and look at 5,700 patient charts, with patients with HIV and hepatitis C. Anybody in IT could say, we have 677 patients with both. But doctors know there’s more to the story. So it turns out another 443 had HIV in the code and hep C mentioned in the notes. Another 14 had hep C in the code, and HIV in the notes.

So using Linguamatics, it’s not 5,700 charts that you need to look at, but 1,150. By using Linguamatics, we narrowed it down to 1,150 patients—those who had both codes. But then we found roughly 460 who had the comorbidity mentioned partly in the notes. Before Linguamatics, all residents had to look at all 5,700 charts, in cases like this one.

So this was a huge time-saver?

Yes, it absolutely 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. And we’re looking at operationalizing this into day-to-day operations. While we’re billing, we can pull files from that day and say, here’s a common co-morbidity—HIV and hep C, with hep C mentioned in those notes—and is there a missed opportunity to get the discrete fields correct?

Essentially, then, you’re making things far more accurate in a far more efficient way?

Yes, this involves looking at patient trials on the research side, while on the clinical side, we can have better quality of care, and more updated billing, based on more accurate data management.

When did this initiative begin?

Well, we’ve been working with Linguamatics for six or seven years. Initially, our work was around discrete fields. The other type of note we look at has to do with text. We had our rheumatology department, and they wanted to find out which patients had had particular tests done—they’re looking for terms in notes… When a radiologist does a report on your x-ray, it’s not like a test for diabetes, where a blood sugar number comes out; x-rays are read and interpreted. The radiologists gave us key words to search for, sclerosis, erosions, bone edema. There are about 30 words. They’re looking for patients who have particular x-rays or MRIs done, so that instead of looking for everyone who had these x-rays done, roughly 400 had these terms. We reduced the number who were undergoing particular tests. The rheumatology department was looking for patients for patient recruitment who had x-rays done, and had these kinds of findings.

So the rheumatology people needed to identify certain types of patients, and you needed to help them do that?

Yes, that’s correct. Now, you might say, 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. I use their ontology, and it helps us understand the appearance of words within structure. I used to be in telecommunications. When all the voice-over IP came along, there was confusion. You hear “buy this stock,” when the message was, “don’t buy this stock.”

So this makes identifying certain elements in text far more efficient, then, correct?

Yes—the big buzzword is unstructured data.

Have there been any particular challenges in doing this work?

One is that this involves an iterative process. For someone in IT, we’re used to writing queries and getting them right the first time. This is a different mindset. You start out with one query and want to get results back. You find ways to mature your query; at each pass, you get better and better at it; it’s an iterative process.

What have your biggest learnings been in all this, so far?

There’s so much promise—there’s a lot of data in the notes. And I use it now for all my preparatory research. And Drexel is part of a consortium here called Partnership In Educational Research—PIER.

What would you say to CIOs, CMIOs, CTOs, and other healthcare IT leaders, about this work?

My recommendation would be to dedicate resources to this effort. We use this not only for queries, but to interface with other systems. And we’re writing applications around this. You can get a data set out and start putting it into your work process. It shouldn’t be considered an ad hoc effort by some of your current people.

 

 


2019 Southern California San Diego Health IT Summit

Renowned leaders in U.S. and North American healthcare gather throughout the year to present important information and share insights at the Healthcare Informatics Health IT Summits.

April 23 - 24, 2019 | Southern California


/article/analytics/drexel-university-moves-forward-leveraging-nlp-improve-clinical-and-research
/blogs/joe-marion/analytics/ai-imaging-where-s-bang-buck

AI in Imaging: Where’s the Bang for the Buck?

January 23, 2019
| Reprints

Over the past year much has been written about the capability of Artificial Intelligence (AI), and what it will mean for imaging services.  At last year’s RSNA, AI was the featured topic and received the lion’s share of publicity. 

The glamorous aspect of AI and Machine Learning has been how AI can assist the radiologist with diagnosis of imaging studies.  A key area of focus has been in chest imaging (https://www.auntminnieeurope.com/index.aspx?sec=sup&sub=aic&pag=dis&ItemID=616828) where there has been some success in triaging abnormal chest images.  The upside of such applications is improved diagnostic efficiency, particularly as healthcare moves toward value-based care.  The downside is that such algorithms require substantial amounts of data to validate, and they will need to go through the FDA approval process, which will take time before they can be fully implemented.

Ultimately, AI imaging applications will pay off.  But, what about the other potentially less-glamorous aspect of applying AI/Machine Learning to the diagnostic process?  By that, I am referring to its use in terms of workflow orchestration.  Aside from interpreting imaging content, AI/Machine Learning applied to workflow orchestration can provide valuable information and assistance in preparing a case for interpretation. 

Take for example Siemens Healthineer’s AI-Rad Companion application (https://www.healthcare.siemens.com/infrastructure-it/artificial-intelligence/ai-rad-companion).  The application provides automated identification, localization, labeling and measurements for anatomies and abnormalities.  Such a capability can improve the radiologist’s efficiency without necessarily employing an algorithm to assess the image.

Other workflow applications can assess the study and mine relevant information from the EHR to present to the radiologist, again with the goal of improving their efficiency and efficacy.  Still other applications match radiologist reading assignments with available studies to improve reading efficiency.  In another twist, one company has demonstrated a capability to further analyze cases, using AI to assign the next appropriate case to a radiologist without the need for a worklist. 

As healthcare providers consolidate, there is a growing need for improvement in resource utilization across facilities.  Smart worklists that can present cases to individual radiologists across facilities can improve the overall efficiency and efficacy of interpretation.  Rule sets that address radiologist availability, reading sub-specialties, location, etc. can help “level-load” reading resources. 

My point is that while AI applications that manipulate images may hold great promise for the future of diagnosis, areas such as workflow orchestration might offer more immediate results in an environment of changing healthcare.  Providers should take a close look at these applications to assess whether they can achieve a more immediate impact on imaging operations.

More From Healthcare Informatics

/news-item/analytics/national-library-medicine-creating-scientific-director-position

National Library of Medicine Creating Scientific Director Position

January 23, 2019
by David Raths, Contributing Editor
| Reprints
New position will oversee Lister Hill National Center for Biomedical Communications and National Center for Biotechnology Information

As part of a reorganization of its intramural research activities, the U.S. National Library of Medicine (NLM) has launched a search for a scientific director. The scientific director will oversee a group of 150 scientific personnel, developing new approaches to data science, biomedical informatics, and computational biology.

In a blog post on the library’s website, director Patti Brennan, R.N., Ph.D., called the move a big step in revving up its intramural research operation.

One of the 27 Institutes and Centers of the National Institutes of Health (NIH), NLM creates and hosts major digital resources, tools, and services for biomedical and health literature, data, and standards, sending 115 terabytes of data to five million users and receiving 15 terabytes of data from 3,000 users every weekday.

NLM’s strategic plan for 2017-2027 positions it to become a platform for biomedical discovery and data-powered health. NLM anticipates continued expansion of its intramural research program to keep pace with growing demand for innovative data science and informatics approaches that can be applied to biomedical research and health and growing interest in data science across the NIH.

A Blue Ribbon Panel recently reviewed NLM’s intramural research programs and recommended, among other things, unifying the programs under a single scientific director. That shift also aligns the library with NIH’s other institutes and centers, most of which are guided by one scientific director.

NLM’s  intramural research program includes activities housed in both the Lister Hill National Center for Biomedical Communications (LHC) and the National Center for Biotechnology Information (NCBI). The researchers in these two centers develop and apply computational approaches to a broad range of problems in biomedicine, molecular biology, and health, but LHC focuses on medical and clinical data, while NCBI focuses on biological and genomic data.

But the Blue Ribbon Panel noted that the boundaries between clinical and biological data are dissolving, and the analytical and computational strategies for each are increasingly shared. “As a result, the current research environment calls for a more holistic view of biomedical data, one best served by shared approaches and ongoing collaborations while preserving the two centers’ unique identities, wrote Brennan, who came to NIH in 2016 from the University of Wisconsin-Madison, where she was the Lillian L. Moehlman Bascom Professor at the School of Nursing and College of Engineering.

She added that having a single scientific director should lead to a sharper focus on research priorities, fewer barriers to collaboration, the cross-fertilization of ideas and the optimization of scarce resources.

The new scientific director will be asked to craft a long-range plan that identifies research areas where the NLM can best leverage its unique position and resources. We’ll also look for ways to allocate more resources to fundamental research while streamlining operational support. “Down the road, we’ll expand our research agenda to include high-risk, high-reward endeavors, the kinds of things that raise profound questions and have the potential to yield tremendous impact,” she wrote.

Besides the scientific director, the NLM is also recruiting three investigators to complement its strengths in machine learning and natural language processing.

 

 

 

 

Related Insights For: Analytics

/news-item/analytics/survey-digital-ai-top-priorities-2019-ehrs-will-dominate-it-spend

Survey: Digital, AI Top Priorities in 2019, but EHRs Will Dominate IT Spend

January 22, 2019
by Heather Landi, Associate Editor
| Reprints

Digital, advanced analytics, and artificial intelligence (AI) are top spending priorities for healthcare executives in 2019, but electronic health record (EHR) systems will dominate technology spending budgets, according to a recent technology-focused healthcare survey.

Damo Consulting, a Chicago-based healthcare growth and digital transformation advisory firm, surveyed technology and service provider executives and healthcare enterprise executives about how the demand environment for healthcare IT is changing and will impact the industry in the coming year. Damo Consulting’s third annual Healthcare IT Demand Survey also analyzes the challenges for healthcare organizations and the perceived impact of macro-level changes.

The report indicates technology vendors will continue to struggle with long sales cycles as they aggressively market digital and AI. For the second year in a row, the rise of non-traditional players such as Amazon and Google will have a strong impact on the competitive environment among technology vendors while EHR vendors grow in dominance.

Among the key findings from the survey, IT budgets are expected to grow by 20 percent or more, with healthcare executives indicating they are more upbeat about IT spend growth than vendors. All the healthcare executives who participated in the survey said digital transformation initiatives are gaining momentum in their enterprises.

However, the majority (75 percent) agree that rapid change in the healthcare IT landscape makes technology decisions harder and only 58 percent believe there are plenty of viable and ready-to-deploy solutions available today in emerging technologies such as AI and digital health solutions. Seventy-one percent agree that federal government policies have provided a boost to healthcare IT spend this past year.

Top IT priorities for healthcare enterprise executives in 2019 are digital, advanced analytics and AI. Of the survey respondents, 79 percent said accelerating digital health initiatives was a top priority and 58 percent cited investing in advanced analytics and AI capabilities as top priorities. However, modernizing IT infrastructure (25 percent) and optimizing EHRs (21 percent) are also significant priorities.

Technology vendors also see AI, advanced analytics and digital transformation as top areas of focus for next year, as those areas were cited by 75 percent and 70 percent of technology and service provider executives, respectively. Thirty-three percent of those respondents cited EHR optimization and 25 percent cited cybersecurity and ransomware. Thirteen percent cited M&A integration as a top area of focus in 2019.

However, EHR systems will dominate technology spending budgets, even as the focus turns to digital analytics, the survey found. Technology and service provider executives who participated in the survey identified EHR system optimization and cybersecurity as significant drivers of technology spend in 2019. Sixty percent of respondents said enterprise digital transformation and advanced analytics and AI would drive technology spend this year, but 38 percent also cited EHR optimization and cybersecurity/ransomware. One executive survey respondent said, “For best of breed solutions, (the challenge is) attracting enough mindshare and budget vs. EHR spends.”

When asked what digital transformation means, close to half of healthcare executives cited reimaging patient and caregiver experiences, while one quarter cited analytics and AI and 17 percent cited automation. As one executive said, “The biggest challenge for healthcare in 2019 will be navigating tightening margins and limited incentives to invest in care design.”

Healthcare executives are divided on whether digital is primarily an IT-led initiative, and are also divided on whether technology-led innovation is dependent on the startup ecosystem.

The CIO remains the most important buyer for technology vendors, however IT budgets are now sitting with multiple stakeholders, the survey found, as respondents also cited the CFO, the CTO, the CMIO and the chief digital officer.

“Digital and AI are emerging as critical areas for technology spend among healthcare enterprises in 2019. However, healthcare executives are realistic around their technology needs vs. their need to improve care delivery. They find the currently available digital health solutions in the market are not very mature,” Paddy Padmanabhan, CEO of Damo Consulting, said in a statement. “However, they are also more upbeat about the overall IT spend growth than their technology vendors.”

Looking at the technology market, healthcare executives perceive a lack of maturity in technology solution choices for digital initiatives, as well as a lack of internal capabilities for managing digital transformation. In the survey report, one executive said, “HIT architecture needs to substantially change from large monolithic code sets to an API-driven environment with multiple competing apps.”

A majority of healthcare enterprise executives view data silos and lack of interoperability as the biggest challenges to digital transformation. And, 63 percent believe the fee-for-service reimbursement model will remain the dominant payment model for the foreseeable future.

In addition, cybersecurity issues will continue to be a challenge for the healthcare sector in 2019, but not the biggest driver of technology spending or the top area of focus for health systems in the coming year, according to the survey.

Healthcare executives continue to be confused by the buzz around AI and digital and struggle to make sense of the changing landscape of who is playing what role and the blurred lines of capabilities and competition, according to the survey report. When asked who their primary choice is when looking for potential partners to help with digital transformation, 46 percent of healthcare executives cited their own internal IT and innovation teams, 17 percent cited their EHR vendor and 8 percent cited boutique consulting firms. A quarter of respondents cited “other.”

For technology vendors, the biggest challenge is long cycles, along with product/service differentiation and brand visibility.

The rise of non-traditional players, such as Amazon, Apple, and Google, will have a strong impact on the competitive healthcare technology environment, the survey responses indicated. At the same time, deeply entrenched EHR vendors such as Epic and Cerner will grow in dominance.

 

See more on Analytics

agario agario---betebet sohbet hattı betebet bahis siteleringsbahis