Enhancing IoMT Security with Explainable Machine Learning: A Case Study on the CICIOMT2024 Dataset

Explainability is critical for medical device security: The paper demonstrates that in healthcare settings, AI systems detecting cyberattacks on medical devices must be transparent and interpretable, not just accurate. Cybersecurity professionals need to understand why an AI flagged a threat to ensure patient safety and maintain trust in the system.

Complex AI models require explanation tools to be trustworthy: While advanced machine learning techniques like Random Forest and CatBoost are effective at detecting sophisticated attacks on connected medical devices, their complexity makes them "black boxes." The study shows that applying explanation techniques (SHAP and LIME) is necessary to make these powerful tools understandable to human decision-makers.

Medical IoT devices face growing security threats: The paper emphasizes that cyberattacks targeting Internet of Medical Things (IoMT)—connected healthcare devices—are rapidly increasing and pose direct risks to patient safety and privacy, making advanced AI-driven security solutions increasingly urgent.

Feature importance insights enable better security decisions: By using explainable AI methods to identify which factors most strongly indicate a cyberattack, the research helps security teams understand the key warning signs of threats, enabling more informed and confident responses to potential breaches.

Responsible AI requires both performance and interpretability: This work illustrates that responsible AI in high-stakes domains like healthcare cannot rely on accuracy alone; systems must balance detection capability with transparency so that human experts can validate, understand, and ultimately be accountable for security decisions.