An Efficient CNN-Based Deep Learning Model to Detect Malware Attacks (CNN-DMA) in 5G-IoT Healthcare Applications

Area of Science:

• Cybersecurity
• Artificial Intelligence
• Health Informatics

Background:

• E-health applications rely on 5G-IoT, increasing vulnerability to cyber threats targeting sensitive patient data stored in the cloud.
• Traditional security measures are insufficient against sophisticated malware attacks in cloud-based e-health systems.
• Deep learning offers potential for advanced threat detection in e-health environments.

Purpose of the Study:

• To propose a novel deep learning model, CNN-DMA, for detecting malware in e-health applications.
• To enhance the security of sensitive patient data within cloud-based healthcare systems.
• To evaluate the effectiveness of the proposed CNN-DMA model against known malware.

Main Methods:

• A hybrid deep learning model, CNN-DMA, was developed using a Convolutional Neural Network (CNN) classifier.
• The model incorporates Dense, Dropout, and Flatten layers, trained with a batch size of 64 and 20 epochs.
• Input images of 32x32x1 dimensions were utilized for the initial convolutional layer, trained on the Malimg dataset with 25 malware families.

Main Results:

• The CNN-DMA model demonstrated high efficacy in detecting malware, specifically identifying Alueron.gen!J.
• The proposed model achieved an accuracy rate of 99% in malware detection.
• Performance was validated against state-of-the-art techniques, confirming its robustness.

Conclusions:

• The CNN-DMA model presents a highly accurate and effective solution for malware detection in e-health applications.
• This deep learning approach significantly enhances the security of cloud-stored patient data.
• The findings support the integration of advanced AI techniques for securing critical healthcare infrastructure.