OCR Data Extraction: Converting Scanned Documents into Machine Readable Information

ocr data extraction

Optical Character Recognition (OCR) in its simplest form is the process whereby a paper document is scanned and the image is captured for processing data into a specific application, such as paper-based EOBs. The text is extracted from an image and converted into machine readable data (as 835 electronic remittance advices) . The problem with OCR processing for paper EOBs, more specifically, is they come from different health insurance payers utilizing their own unique document templates. Therefore, not all of the data is in the same ‘mapped’ locations across all payer EOBs.

With standardized data elements, where the fields and locations are all consistent, additional programming features can be added to OCR technology to streamline the processing and validation of information.

RCMS has decades of Optical Character Recognition experience, integrated with Document Management/ECM platforms and rules-based, machine learning programming capabilities.

Once mapping and processing rules have been established for a specific payer, the RCMS Advanced OCR system will check for certain types of characters (or number of characters) in specific alpha-numeric sequences — before human intervention is required. Any extracted data that doesn’t meet certain criteria is then flagged for review and edits — with staff having access to the original scanned image. Interfaces to other HIS applications can also be utilized to look for and correct patient information.

RCMS OCR Technology Features

  • Optical Character Recognition with Artificial Intelligence (AI)
  • Paper EOB & Insurance Claims Processing Expertise
  • Document Classification & Indexing
  • Data Extraction & Normalization
  • Key from Image (KFI) Functionality (for QA review and corrections)
  • Document Management Content Services using OnBase ECM platform
  • Intelligent Character Recognition (ICR) Learning Capabilities
  • Data Warehousing & Analytics

Advanced OCR/ICR Cloud-Based or On-Premise Deployment

RCMS provides for hosted or on-site Advanced OCR/ICR installations, combined with document management solutions for a variety of high-volume business applications — ranging from Accounts Payable and HR to Legal Departments. Intelligent character recognition (ICR) is also utilized to process handwritten text and to allow for the software to learn different font styles. The key to OCR success is in having the right volume of similar documents needing to be processed on a regular basis.

With a significant number of payers (or payors) still not sending electronic 835 files, RCMS can help you to convert paper EOBs to HIPAA-compliant 835 files, with line item detail, for processing. We can also contract with a business process outsourcing partner, such as ReMedics LLC, to outsource all your 835 conversion and payment processing needs.

Contact RCMS about how our Advanced OCR technology can work for your organization.

Advanced OCR/ICR is a trademark of RCMS, LLC