Linking AI Initiatives to Real-World Outcomes

Use Case

CLAIM ADJUDICATION: AUTOMATED  OVERPAYMENT & RECOVERY

 
Client

Leading end-to-end payment integrity organization based in Indiana.

Problem

The client offers Claim Overpayment & Recovery solutions. They partner with their customers to identify, eliminate and recover overpayments.

The client wanted to automate the adjudication (bill review) process by classifying each item description into a charge type accurately in order to apply business rules of adjudication.

Methodology
  • Combination of Deep Learning based professional service and modularized AI components used according to organization’s cognitive maturity
  • Business transformation with smarter cognitive outcomes
  • Adding industry, business and customer insights to the value chain
  • Seamlessly integrate data across functions
Solution
  • Natural Language Processing and advanced Machine Learning algorithms used to understand the context of the item description with a continuous learning system classifying each item into a charge type (e.g. laboratory changes, monitoring charges etc.)
  • Workflow consists of a classification engine which does: pre-processing, concept identification, and advanced machine learning to classify each item to a charge class with a confidence score
Benefits:
  • Generated $2.7M in savings
  • 40% Increased assessment speed
  • 90% accuracy with minimal supervision
  • Processing costs have been reduced immensely
 
 

© Abzooba 2021. All rights reserved  

© Copyright www.aiinc.cloud