Answers are grounded in this paper and cite their sources.
Comprehensive review identifies best practices and challenges for deep learning in IoMT security.
The Internet of Medical Things (IoMT) is transforming modern healthcare by interconnecting medical sensors, devices, and clinical systems. It enables real-time data collection, remote patient monitoring, personalized treatment, and reduced hospital stays, while also supporting data-driven research in the medical domain. However, the rapid expansion of IoMT infrastructure has introduced substantial security vulnerabilities, making these systems highly susceptible to sophisticated cyberattacks. This study presents a comprehensive and rigorous survey of intrusion detection techniques for IoMT systems using deep learning. The primary objective is twofold. First, we systematically review state-of-the-art deep learning–based approaches across all phases of the machine learning project life cycle, including data collection, exploration and preprocessing, feature engineering, model selection and training, evaluation, deployment, and maintenance. This holistic perspective provides beginner researchers with clear guidance on best practices for designing and implementing intrusion detection systems for IoMT environments while also highlighting key research gaps. Second, we identify critical challenges at each stage of the development pipeline and propose potential solutions to address them. In addition, this survey outlines emerging research directions to support the development of robust, secure, and scalable IoMT intrusion detection systems.
No takes yet. Share an insight, caveat, or question.
Shah et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: