Securing the Digital Frontier: The Role of Technology in Social Medical Public Healthcare Security

Abstract

The rapid expansion of digital connectivity within social medical public healthcare systems (SMPH) has fundamentally transformed the way patient care is delivered. However, it has also made sensitive data vulnerable to a wide range of cybersecurity threats. This study introduces and assesses a new hybrid deep learning model, GANA-AO, with the aim of improving real-time anomaly detection and threat prevention in SMPH. GANA-AO leverages the capabilities of Generative Adversarial Networks (GAN) and Autoencoders, enhanced by Adam optimization, to achieve outstanding accuracy and generalizability. Generative Adversarial Networks (GAN) produce authentic artificial data to supplement the training dataset and tackle the problem of imbalanced classes. On the other hand, Autoencoders acquire compact representations of normal data, aiding in the detection of anomalies by identifying deviations. Adam optimization effectively adjusts model hyperparameters, thereby improving performance. The efficacy of GANA-AO is demonstrated through our experiments conducted on the publicly accessible IoT-23 dataset. The model demonstrates an exceptional accuracy of 98.33% and a True Positive Rate (TPR) of 98.67%, surpassing the performance of baseline models by a significant margin. The results emphasize the capability of GANA-AO to enhance SMPH cybersecurity by promptly detecting and addressing malicious activities, protecting sensitive healthcare data, and ensuring patient safety. This paper not only introduces a robust technical solution but also highlights the vital significance of technology in safeguarding the digital boundaries of SMPH. By adopting cutting-edge approaches such as GANA-AO, we can establish a stronger and more adaptable system, promoting confidence and enabling patients in the digital era of healthcare.

DOI: https://doi.org/10.52710/seejph.487