Natural Language Processing of Clinical Documents

As organizations across the care continuum look to drive a greater understanding of patients and populations, the ability to leverage alternate data sources is growing. This opportunity has led many organizations to evaluate the information in clinical documentation, charts, and electronic medical records. The importance of this text cannot be understated in evaluating treatment protocols, patient diagnosis, and the ability to understand potential missed diagnosis, amongst many other applications.

Healthcare organizations are rapidly trying to understand how to unlock the value from within these documents and leverage the structured and unstructured contents for analytical and operational purposes.

Knowledgent utilizes Natural Language Processing (NLP) technologies to extract pertinent data from clinical documents such as patient charts and Continuity of Care Documents (CCDs). These documents can reside in multiple forms (electronic, PDF, paper, fax, e-mail, etc.) and conform to various standards, including HL7 and FHIR. This data can be leveraged to improve member health, quality ratings, risk-revenue and fraud, waste and abuse. Users can interact with the data both within a searchable user interface as well as within Hadoop.

How It Works

  • The source of medical notes data is identified and imported into a Hadoop environment (HDFS)

  • Text mining, NLP, and parsing methods extract human readable text from clinical documents, such as: patient name, age, gender, and Medical Record Number (Optional segments include: Payers, Lab results, Medication lists, etc.)

  • The clinical documents are converted into viewable HTML and PDFs and structured elements are stored within HDFS

CCD Example

Patient Chart Example

Documents are Accessible via a Searchable UI

Bringing It All Together: The Unified Patient Record

Knowledgent’s Unified Patient Record (UPR) framework is a game-changing enabler for healthcare companies. The UPR provides a “single source of truth” for patient data which enables the ability to perform a wide variety of crucial, data-driven analyses.

The UPR provides a ‘360 degree’ view of each individual patient, including the extracted clinical document data, medical history, claim statuses, MD utilization, Rx, balances, and upcoming/historical appointments scheduled. It also allows the ability to run high-level reports for population health initiatives and patient segmentation.

Sample Analytical Questions Answered with the UPR:

  • How many patients were seen over a time period

  • How many were new patients

  • What types of visits

  • Average collections per visit type, physician, or facility

  • How patient population changed, where it grew

  • Patients’ preferences and where the data is stored

Benefits of Knowledgent’s UPR:

  • Integrated clinical, financial, and member data in a single place

  • Rapidly-ingested data while promoting a single source of truth

  • Enhanced operational reporting

  • Elimination of redundant data and analytical efforts

  • Better, cleaner data to support marketing initiatives or diagnoses-based outreach for patient populations

  • Improved customer service and patient satisfaction

  • Improved employee satisfaction through transparency and unity across the enterprise

Never miss an update:

Subscribe to our newsletter!

Newsletter Sign Up Form

  • This field is for validation purposes and should be left unchanged.

New York, NY • Warren, NJ • Boston, MA • Toronto, Canada
©2018 Knowledgent Group Inc. All rights reserved.