A State-of-the-Art Healthcare Decision Support System

  • Client: A leading insurance company.

  • Objective: To create a decision support system using claims data to minimize healthcare costs.

  • Solution and outcome: Created an integrated API with multi-crawler system along with literature-based data mining and deep learning model. Unstructured and structured datasets were collected from open and proprietary sources, cleaned and integrated into an automatically updated database. Predictive modeling and Natural Language Processing were applied to flag erroneous outputs, suspicious claims and poor cost plan. This helped detect fraud and attract customers by offering healthcare plans that specifically suit their needs.