Hybrid Anomaly Detection in Time Series by Combining Kalman Filters and Machine Learning Models.

Due to connectivity and automation trends, the medical device industry is experiencing increased demand for safety and security mechanisms. Anomaly detection has proven to be a valuable approach for ensuring safety and security in other industries, such as automotive or IT. Medical devices must operate across a wide range of values due to variations in patient anthropometric data, making anomaly detection based on a simple threshold for signal deviations impractical. For example, surgical robots directly contacting the patient's tissue require precise sensor data. However, since the deformation of the patient's body during interaction or movement is highly dependent on body mass, it is impossible to define a single threshold for implausible sensor data that applies to all patients. This also involves statistical methods, such as Z-score, that consider standard deviation. Even pure machine learning algorithms cannot be expected to provide the required accuracy simply due to the lack of available training data. This paper proposes using hybrid filters by combining dynamic system models based on expert knowledge and data-based models for anomaly detection in an operating room scenario. This approach can improve detection performance and explainability while reducing the computing resources needed on embedded devices, enabling a distributed approach to anomaly detection.

Threat Assessment and Risk Analysis (TARA) for Interoperable Medical Devices in the Operating Room Inspired by the Automotive Industry.

Prevailing trends in the automotive and medical device industry, such as life cycle overarching configurability, connectivity, and automation, require an adaption of development processes, especially regarding the security and safety thereof. The changing requirements imply that interfaces are more exposed to the outside world, making them more vulnerable to cyberattacks or data leaks. Consequently, not only do development processes need to be revised but also cybersecurity countermeasures and a focus on safety, as well as privacy, have become vital. While vehicles are especially exposed to cybersecurity and safety risks, the medical devices industry faces similar issues. In the automotive industry, proposals and draft regulations exist for security-related risk assessment processes. The medical device industry, which has less experience in these topics and is more heterogeneous, may benefit from drawing inspiration from these efforts. We examined and compared current standards, processes, and methods in both the automotive and medical industries. Based on the requirements regarding safety and security for risk analysis in the medical device industry, we propose the adoption of methods already established in the automotive industry. Furthermore, we present an example based on an interoperable Operating Room table (OR table).