PROJECT OVERVIEW

Using Advanced Technology in Medical Review

blue-arrow

Health care insurance fraud is a pressing problem, causing substantial and increasing costs in medical insurance programs. Due to large amounts of claims submitted, review of individual claims becomes a difficult task and encourages the employment of automated pre-payment controls and better post-payment decision support tools to enable subject matter expert analysis. Investing in predictive algorithms to craft reliable conclusions about data is becoming a popular and extremely beneficial method in the area of analysis, especially with the many ways to approach data mining.

TECHNOLOGIES
KNIME, a powerfully flexible tool that grants the user full control over workflow creation and deployment is the engine of choice for conducting predictive analysis. There is a wide selection of nodes that are available to the user, who chooses a particular layout for the ordering of these nodes; the GUI’s drag and drop features make it simple to create sophisticated workflows.

METHODOLOGY
Building a KNIME workflow that connects nodes that perform appropriate functions. All nodes have ports, which act as intermediaries and connect streams between nodes to allow data to move between nodes. The tables created by KNIME for the input data travel along these nodes in a user-defined path, and will follow directions exactly how they are declared in the workflow.

PROCESS
Data label preprocessing and replacement, parameterized data filtering, outlier detection with statistical analysis by grouping the input data by the target variable and compute the mean and standard deviation for the numerical variable in question. The user can change the group and aggregation column via the metanode context menu. The lower branch of the workflow is a refinement of this approach and allows the identification of outliers across several variables. KNIME Power BI Integration. KNIME Analytics Platform has a seamless integration with Power BI, which allows you to push KNIME tables into Power BI datasets. To visualize our results, Power BI’s custom visual Sand Dance was chosen. The Sand Dance custom visual puts this innovation from Microsoft Research into the hands of users through the open, extensible visualization framework of Power BI that can leverage rich connectivity, modeling, and interactive reporting.

blue-arrow