The Challenge

  • Identify fraud, waste and abuse in federal Centers for Medicare and Medicaid (CMS) Affordable Care Act Exchange Enrollments
  • Identify fraud, waste and abuse in state healthcare exchange enrollments

The Reveal Solution

Since 2018, Reveal has been contracted by the Centers of Medicare and Medicaid Services (CMS) Center for Program Integrity (CPI) to identify waste, fraud and abuse (FWA) in the Affordable Care Act (ACA) Exchanges using artificial intelligence and machine learning techniques. Reveal has leveraged its proprietary Big Data Processing platform– Reveal Analytics Controller (RAC) – and AI/ML techniques to help identify anomalous records in the ACA enrollments datasets from Federally Facilitated Exchanges. Reveal has provided CMS CPI with an end-to-end solution that ingests, pre-processes, and analyzes enrollment data while providing deep insights into the data through interactive and user-friendly visualization dashboard.

The Reveal FWA solution does the following:

  • Uses numerous data sets that include enrollment, payments, subsidies, member, and complaint details
  • Identifies outliers using average unsupervised fitness value per county
  • Analyzes assistors used, density of use, and use by region using U.S. address data and location-based info
  • Traces services use back to enrollment
  • Detects how elements of ACA Exchanged interact with and relate to others (e.g a rehab facility has been associated with fraud and has fake enrollees)
  • Processes and analyzes data using Reveal Analytics Controller (RAC) and its AI capabilities
  • Provides an ontological reference knowledge base that feeds back into the system and creates a framework to identify and continuously update relationships
  • Uses Alteryx as a platform for data ingestion, analytics, and work flow engine to implement AI algorithms with advanced analytics capabilities
  • Uses Tableau for advanced visualizations of results (dashboards, reports)
  • Continuously refines analysis for ongoing quality improvements

Why Reveal?

Reveal solutions are purpose built to meet specific client needs. Our vendor-neutral solutions are designed to elicit the maximum value possible from your big data, and to simplify and automate big data and analytics processes. Examples of advanced analytics techniques we use include the following:

  • Pattern recognition to identify phishing emails based on content or sender info, identifying malware, etc.
  • Anomaly detection to spot unusual activity, data, or processes such as fraud detection for online banking
  • Natural language processing (NLP) converts unstructured text such as a webpage into structured, actionable intelligence
  • Predictive analytics for processing data and identifying patterns to make accurate predictions and identify outliers
  • Application of AI/ML techniques based on trends and insights to develop a probabilistic/stochastic (time series) model to predict adversary behavior

Create and Curate the Right Algorithms

  • Reveal combines the best ensemble of open source and proprietary unsupervised and supervised algorithms to optimize results
  • We ensure general agreement between algorithms to maximize their impact on results

Dimension Reduction

  • We use Principle Component Analysis (PCA) to flatten the dimensions in the output to improve algorithm speed and performance
  • Reveal’s visualization solution enables easier viewing of outliers

Fitness Values

  • Fitness value is a single number that summarizes the degree to which a record is similar to other records in the dataset
  • Set a fitness value range (-ve #, +ve #) to separate inliers and outliers – outliers have more negative fitness values

Training the Algorithms

  • Assign a Threshold Value (TV) to separate inliers (FV >TV).
  • Separated the rows in inliers/outliers using the unsupervised FV. These labeled rows are used to train the Supervised ML Algorithms