A Blockchain Enabled Predictive, Analytical Model for Fraud Detection in Healthcare Data

Abstract

With technological advancements, existing Electronic Health Record (EHR) management systems, along with Machine Learning (ML) algorithms, are geared up for a huge transformation in the healthcare business. Healthcare data is created every microsecond from a variety of sources and contains a wealth of information. However, there are a number of problems with the data in these systems, including security, dependability, and accessibility, to mention a few. Consequently, there is always a risk associated with such data. Conventional methods are ineffective in addressing these crucial issues because they lack a uniform structure for data security and availability strategies. To safeguard the integrity of the use of EHR system, a new solution is needed to manage government security regulations and improve data accessibility. Furthermore, past healthcare data (healthcare provider data, inpatient/outpatient data) must be analyzed using machine learning approaches to predict resource requirements and probable fraud, which is critical for driving better future decisions. This study intends to bridge the gap by offering a prototype as architecture for a secure and intelligent EHR management system that relies on blockchain technology and machine learning methods. Finally, the proposed system is evaluated by considering transaction latency and throughput using the Hyperledger Caliper tool. The accuracy, precisionrecall, and F1 score are the performance metrics used to evaluate prediction models.