Intelligent Neighborhood Indexing Sequence Model for Healthcare Data Encoding
Ibrahim M. El‐Hasnony, Mohamed Elhoseny, Mohammed Khaled Hassan · Journal of Intelligent Systems and Internet of Things · 2019
Recently, information security in the healthcare sector has become essential to ensure confidentiality in medical data. At the same time, automated disease diagnosis using deep learning (DL) models also gained considerable attention to accomplish enhanced classification performance. This paper designs an intelligent neighborhood indexing sequence based on encoding with a classification model for healthcare information security (INISEC-HIS). The proposed INISEC-HIS technique aims to accomplish security in medical data transmission and diagnosis. The neighborhood indexing sequence (NIS) technique is applied to securely transmit the data, which transforms the medical data into an encoded format. Besides, a novel artificial fish swarm algorithm (AFSA) with deep neural networks (DNN) model is used for the classification process. The design of AFSA to optimally adjust the hyperparameters of the DNN model shows the study's novelty. An extensive simulation analysis takes place to examine the improved outcomes of the INISEC-HIS technique, and the obtained results highlighted the supremacy over the other techniques.