Premium Reports
Natural Language Processing

Natural Language Processing
Glimpses Into the Future of Unstructured Data Mining

Authored by: | Read Time: 3 minutes

In many ways, the current healthcare climate has created the perfect storm for a surge in provider mindshare toward leveraging NLP to help improve patient care. However, providers currently face challenges evaluating the different NLP offerings and how they will perform in specific contexts. Additionally, providers note that current NLP solutions face significant challenges in unraveling the complexities of medical language. KLAS spoke with academic facilities, large IDNs, and known NLP users to provide an overview of the NLP market, provider strategies, and NLP use cases and a very early look at what vendors are offering in NLP solutions.

2016 Natural Language Processing Five General Areas

1. PROVIDERS SKEPTICAL ABOUT IBM HYPE

While IBM is the most-mentioned NLP vendor, providers are cautious about IBM's NLP potential; they report that IBM's tools are challenging to use and have limited functionality. While providers acknowledge IBM is working hard to provide NLP capable of decision support, they are concerned about the product's accuracy and its application with complex healthcare language. IBM will face significant challenges in overcoming provider skepticism that Watson will prove to be a competitive option.

2016 Natural Language Processing Depth of Validated NLP Use

“The things Watson did for Jeopardy were very slick, but they do not really extend out to the complexities of medicine. It is not clear to me whether Watson is really going to do what people want. Watson seems better for figuring out the best treatment for a patient or other possible diagnoses for a patient given certain findings. Whereas a lot of people, especially on the research side, have 500 or 5,000 patients with colon cancer, and they want to abstract information about how many have had adverse events, how many have had difficulty breathing, or how many have fallen and fractured their leg after chemotherapy.” CHIO


2. M*MODAL'S AND NUANCE'S FRONT-END CAPABILITIES DRIVING CAPD ENERGY

Providers are demonstrating early interest in M*Modal's and Nuance's computer-assisted physician documentation (CAPD) solutions to complement front-end speech applications at the point of care. The solutions are as yet unproven, but early feedback from providers live with CAPD solutions includes admiration for M*Modal's context-driven NLP engine. Additionally, providers are excited about the benefits of both Nuance's and M*Modal's ability to offer real-time NLP feedback to improve the quality of physician documentation.

2016 Natural Language Processing NLP Use Cases

3. APIXIO, HEALTH FIDELITY, AND LINGUAMATICS OFFER EARLY SUCCESS FOR POPULATION HEALTH USE CASES

Many providers are looking to NLP engines to help drive population health initiatives by mining unstructured data. In early conversations with providers using NLP in this area, Apixio, Health Fidelity, and Linguamatics demonstrate early promise. One provider reported Apixio as having maturing algorithms that are used to address risk adjustment. Health Fidelity is developing a reputation for having an engine that can go beyond open-source tools to parse text at a deep level. Similarly, providers see Linguamatics as an emerging product that has potential for applications that offer real-time clinical decision support.

4. CERNER EXPANDS NLP FUNCTIONALITY WHILE EPIC ALLOWS THIRD-PARTY INTEGRATION

Cerner and Epic customers looking to leverage solutions from their core vendors will find that current NLP solutions are focused on front-end documentation. Cerner’s NLP solution allows providers to mine EMR data for patient summaries, coding reconciliation, and regulatory use cases. Epic’s NoteReader function is receiving consideration for providing an interface to allow third-party NLP solutions to mine Epic EMR data. While Epic currently focuses on allowing users to partner with outside vendors for NLP needs, providers express optimism about Cerner’s willingness to continue to expand NLP capabilities for future use cases.

2016 Natural Language Processing Depth of Validated NLP Use

"We are very interested in NLP. I feel that a lot of places are not focusing on the use case in NLP. They are trying to build the best hammer and then find a nail. Our first task is to figure out where our needs are and what our strategies are for ACOs and population health, and then we need to see what the best tool is for our needs.” CHIO


2016 Natural Language Processing Provider Paths Through Progressive NLP Use

Want to see full details?

KEY TOPICS

  1. Providers Skeptical About IBM Hype
  2. M*Modal’s and Nuance’s Front-End Capabilities Driving CAPD Energy
  3. Apixio, Health Fidelity, and Linguamatics Offer Early Success for Population Health Use Cases
  4. Cerner Expands NLP Functionality While Epic Allows Third-Party Integration
 Download Report Brief  Download Full Report

This material is copyrighted. Any organization gaining unauthorized access to this report will be liable to compensate KLAS for the full retail price. Please see the KLAS DATA USE POLICY for information regarding use of this report. © 2018 KLAS Enterprises, LLC. All Rights Reserved. NOTE: Performance scores may change significantly when including newly interviewed provider organizations, especially when added to a smaller sample size like in emerging markets with a small number of live clients. The findings presented are not meant to be conclusive data for an entire client base.