Feature optimization using Hybrid Metaheuristic Red Deer and Dragonfly Algorithms for Multi-disease prediction

In health informatics, healthcare systems are developing Electronic Medical Records (EMR) by compiling a wide range of patient-related data. The adoption of EMR has potential to revolutionize healthcare by raising the bar for the standard of treatment all patients receive. Electronic medical records introduction could lead to a rise in the quantity of available clinical data. The processing of this massive data will aid in both the discovery of new illness patterns and the provision of individualized care for patients. The study of health risks and their repercussions can benefit greatly from the use of machine learning techniques on EMR datasets. The approach builds on EMR that accurately diagnose and predict various types of diseases by developing efficient algorithm and method for the processing and analysis of biomedical data using machine learning approach to provide better performance and achieve high accuracy. An ensemble-based machine learning Hybrid Disease Optimization (HDO) method is proposed by employing Hybrid Metaheuristic Red Deer Algorithm (RDA) and Dragonfly Algorithm (DA) Optimization Algorithms to boost performance of the model. The comparison of all approaches is studied and concluded that, proposed HDO method of developing predictive machine learning model using electronic medical records yields significantly improved results than state-of-art models.