Detection of Physiological Data Tampering Attacks with Quantum Machine Learning

2025-02-09

AI Summary

Physiological time-series data from cloud-based medical devices and wearable sensors face growing security threats from white-box adversarial manipulations—specifically, label-flipping data poisoning and adversarial perturbations. Method: This paper presents the first systematic evaluation of quantum machine learning (QML) for detecting such tampering, proposing a unified detection framework integrating quantum state encoding, quantum support vector machines (QSVM), and quantum neural networks (QNN), benchmarked against classical models including LSTM and random forests. Contribution/Results: Experimental results demonstrate that QML achieves 75–95% detection accuracy against label-flipping poisoning attacks—significantly outperforming classical baselines—and 45–65% accuracy against adversarial perturbations, also surpassing conventional approaches. These findings reveal QML’s intrinsic capacity for modeling high-dimensional, nonlinear physiological signals and extracting robust features, establishing a novel quantum-enhanced paradigm for securing healthcare data.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Adversarial Attacks & Robustness
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

The widespread use of cloud-based medical devices and wearable sensors has made physiological data susceptible to tampering. These attacks can compromise the reliability of healthcare systems which can be critical and life-threatening. Detection of such data tampering is of immediate need. Machine learning has been used to detect anomalies in datasets but the performance of Quantum Machine Learning (QML) is still yet to be evaluated for physiological sensor data. Thus, our study compares the effectiveness of QML for detecting physiological data tampering, focusing on two types of white-box attacks: data poisoning and adversarial perturbation. The results show that QML models are better at identifying label-flipping attacks, achieving accuracy rates of 75%-95% depending on the data and attack severity. This superior performance is due to the ability of quantum algorithms to handle complex and high-dimensional data. However, both QML and classical models struggle to detect more sophisticated adversarial perturbation attacks, which subtly alter data without changing its statistical properties. Although QML performed poorly against this attack with around 45%-65% accuracy, it still outperformed classical algorithms in some cases.

Problem

• Detecting physiological data tampering
• Evaluating Quantum Machine Learning effectiveness
• Comparing QML and classical models

Innovation

• Quantum Machine Learning
• Detects Physiological Data Tampering
• Outperforms Classical Algorithms