Defending Medical AI from Potential Cyber Threats is the Focus for Scientists
SecureNeuroAI: Protecting AI-Based Medical Devices from Cyberattacks
The University of Bonn, University Hospital Bonn, and FIZ Karlsruhe have joined forces in a project called SecureNeuroAI, with the aim of safeguarding AI-based medical devices and applications from cyberattacks. This initiative is crucial in ensuring the reliability and resilience of critical medical technologies, particularly those involving neuro-AI applications, against cybersecurity threats that could compromise patient safety and medical treatment efficacy.
At the helm of the project is Elena Demidova, the head of the "Data Science and Intelligent Systems" (DSIS) working group at the University of Bonn, who serves as the coordinator of SecureNeuroAI. She is supported by a team of experts from various disciplines, including computer science, medical research, and information engineering.
The research department Immaterial Rights (IGR) led by Franziska Boehm at FIZ Karlsruhe is responsible for analyzing data protection and IT regulations, as well as the legal issues of artificial intelligence. Michael Meier from the University of Bonn brings expertise in current IT security topics, focusing on the vulnerabilities of networked medical devices and their accompanying infrastructures.
The UKB plays a significant role in the project, collecting multimodal data for seizure detection and preparing it under clinical conditions to create a realistic data basis for the AI models. Bjoern Kruger, from the UKB, emphasizes the importance of secure system thinking in AI-based medical applications, especially when dealing with sensitive patient data.
The project team is also working on defining technical, organizational, and legal measures to support the integration of the developed AI methods into clinical and home applications. The goal is to create a technological basis that improves the integrity, availability, and reliability of AI-based medical devices, increasing their resistance to cyberattacks.
One of the key objectives of the SecureNeuroAI project is to develop data authentication procedures for AI-supported methods to detect medical emergencies, using epileptic seizures as an example. The project aims to address the challenge of manipulation recognition, a significant issue in the field of AI-assisted medical emergency detection due to the complexity of data patterns and limited data availability.
The project is funded by the Federal Ministry of Research, Transfer, and Space (BMFTR) with approximately 2.5 million euros over three years. The project involves cyber-secure AI models that analyze multimodal sensor data, including vital parameters such as heart and breathing rates.
The results of the SecureNeuroAI project could find application beyond the detection of epileptic seizures. Julia Klinkusch, a freelance journalist specializing in science and health topics, has contributed to the article. Klinkusch's areas of focus include climate, AI, technology, environment, and medicine/medical technology.
In summary, the SecureNeuroAI project is a significant step towards ensuring the safety and efficacy of AI-driven medical devices in the face of emerging cyber threats. By developing robust security frameworks for AI in healthcare settings, the project contributes to safer medical AI innovation and compliance with stringent safety requirements in healthcare.
- The SecureNeuroAI project, led by Elena Demidova, aims to develop data authentication procedures for AI-supported methods that detect medical emergencies, such as epileptic seizures, to mitigate the challenge of manipulation recognition due to the complex data patterns and limited data availability.
- The project's goal is to create a technological basis that improves the integrity, availability, and reliability of AI-based medical devices, increasing their resistance to cyberattacks, and contributing to health-and-wellness and medical-conditions sectors by ensuring safer medical AI innovation and compliance with stringent safety requirements.
- Collaborating with the University Hospital Bonn and FIZ Karlsruhe, the project involves cyber-secure AI models that analyze multimodal sensor data, including vital parameters such as heart and breathing rates, and explores the intersection of cybersecurity and artificial-intelligence in the field of technology and medical-conditions, promoting the advancement of health-and-wellness and the medical-industry.
