Improving internet of health things security through anomaly detection framework using artificial intelligence driven...

Area of Science:

• Cybersecurity
• Health Informatics
• Artificial Intelligence

Background:

• Healthcare systems face increasing cybersecurity threats due to widespread technology adoption.
• Internet of Health Things (IoHT) devices are vulnerable to cyberattacks, necessitating robust detection methods.
• Machine learning (ML) and Artificial Intelligence (AI) offer advanced capabilities for anomaly detection in cybersecurity.

Purpose of the Study:

• To develop an AI-driven technique for detecting cyberattacks in the Internet of Health Things (IoHT) environment.
• To enhance the security of IoHT by accurately classifying anomalous patterns indicative of cyber threats.
• To improve the efficiency and accuracy of cyberattack detection in healthcare settings.

Main Methods:

• The Enhancing Internet of Health Things Security through Cyberattack Detection Using Serial Exponential Golf Optimization (EIoHTSCD-SEGO) technique was developed.
• Data preprocessing involved TF-IDF for feature vectors and min-max normalization.
• An ensemble of deep learning (DL) classifiers (RNN, BiLSTM, KELM) was optimized using the Serial Exponential Golf Optimization Algorithm (SEGOA).

Main Results:

• The EIoHTSCD-SEGO technique achieved a superior accuracy of 99.33% in cyberattack detection.
• The method demonstrated effective classification of anomaly detection using AI-based data science.
• Performance validation was conducted using the ECU-IoHT benchmark dataset.

Conclusions:

• The developed EIoHTSCD-SEGO technique significantly improves cybersecurity in IoHT environments.
• AI-driven anomaly detection, particularly with DL ensembles and optimization algorithms, is highly effective for healthcare cybersecurity.
• The study highlights the potential of advanced ML/AI techniques to mitigate cyber risks in connected healthcare systems.