Securing Healthcare Data in Mobile Edge Computing: A Hybrid Deep Learning Framework

Strategic Insights -

A new study by researchers at the College of Applied Sciences, AlMaarefa University, has made significant strides in developing a hybrid deep learning framework that combines multiple machine learning strategies to preserve privacy and detect anomalies in healthcare data. The research proposes a novel technique, LDExGRNN_SVA-AdBCR, which integrates differential privacy techniques with XGBoost and deep Q-learning reinforcement strategies to achieve privacy preservation.

Key Takeaways:

  • The LDExGRNN_SVA-AdBCR method provides a scalable solution for detecting anomalies in MEC environments and securing healthcare data.
  • The technique integrates multiple machine learning strategies, including differential extreme gradient reinforcement neural network, stochastic vector autoencoder-based adversarial Bayes convolutional regression, and Naïve Bayes, support vector machines, generative adversarial networks, and vector autoregression.
  • The experimental analysis was evaluated on various healthcare datasets, achieving a privacy rate of 98.1%, prediction accuracy of 98.75%, scalability of 98.3%, and F1-score of 98.34%.
  • The LDExGRNN_SVA-AdBCR method outperformed other existing methods in terms of privacy preservation and anomaly detection.
  • The research provides a novel solution for addressing the challenges of data privacy and anomaly detection in MEC environments.

Statistics:

  • The LDExGRNN_SVA-AdBCR method achieved a privacy rate of 98.1%.
  • The method achieved a prediction accuracy of 98.75% on various healthcare datasets.
  • The scalability of the method was 98.3%.
  • The F1-score of the method was 98.34%.

Sources:

  • Securing Healthcare Data In Mobile Edge Computing: a Hybrid Deep Learning Framework for Privacy and Anomaly Detection. Knowledge and Information Systems, 2025.
  • Anndh Sam Chandra Bose, et al. Securing Healthcare Data in Mobile Edge Computing: A Hybrid Deep Learning Framework. Journal of Engineering, October 20, 2025; p 4302.