Big data analytics in healthcare environment using chaotic red deer optimizer with deep learning for disease classification model

Deep learning (DL) and big data analytics are powerful tools when combined for disease detection and healthcare applications. They enable the processing of vast amounts of medical data to uncover patterns, make predictions, and assist healthcare professionals in diagnosing diseases and improving patient care. Existing disease detection model often faces two challenging issues namely feature selection and insufficient hyperparameter tuning, which considerabley affects the performance and generalization abilities of DL models in disease detection. To resolve these issues, this article focuses on the design of a new Chaotic Red Deer Optimizer with Deep Learning based big data analytics for Disease Classification (CRDODL-BDADC) technique in the healthcare environment. The CRDODL-BDADC technique resolves the high dimensionality problem by the use of CRDO based feature selection approach. For disease detection process, convolutional bidirectional gated recurrent unit (CBGRU) is applied for disease diagnosis and its hyperparameters are selected by using Seeker Optimization Algorithm (SOA). Furthermore, Map Reduce is applied to manage healthcare BD. A widespread set of simulation outcomes occur on the PCOS database, from the Kaggle repository to illustrate the greater performance of the CRDODL-BDADC technique. The simulation values highlighted the enhanced efficiency of the CRDODL-BDADC method over other recent approaches under various measures.