An Effective Graph Database-Centric Patient Healthcare Data Management Using A Robust HGCRN and LIM2DCE

Abstract

In basic, effective data management in the healthcare domain is a difficult task due to the complex relationship between the patients and the healthcare sector. However, none of the prevailing models focused on analyzing the complex relationships and hidden patterns between the client data for identifying the intention of the patients. Therefore, this article proposes an effective graph database-grounded patient healthcare data management using a robust HGCRN and LIM2DCE. Firstly, the patients are registered into the healthcare applications by entering their details, followed by privacy preservation. Subsequently, a patient’s unique identifier is generated regarding the patient’s information. Next, the patient logs in to the application, and then the healthcare data is encrypted to ensure data privacy. Here, the encrypted data is subjected to the proposed readmission classification model. Initially, the historical dataset is pre-processed. Thereafter, the outlier detection and elimination are done, followed by dimensionality reduction. Next, the graph is constructed by using the neo4j bloom. Also, the client is segmented according to the similar information. Further, the nodes and edges are extracted and then inputted to the proposed HGCRN, where the readmission prediction is done. Accordingly, the intention of the patients is described via the LIM2DCE. The proposed work had an impressive performance with 98.4578% accuracy.