Closed-loop medical devices interact with patient physiologcy in closed-loop, therefore the safety and efficacy of the closed-loop devices should be evaluated within their physiological context. The biggest challenge for model-based closed-loop validation is the lack of appropriate models of the physiological context. In this work we developed a heart model structure which can be used for closed-loop validation of implantable cardiac devices.
Closed-loop model checking of medical devices requires model(s) of the physiological context which not only general enough to capture the behaviors of physiological conditions specified in the requirement, but also expressive enough to explain the physiological phenomena. In this project we use implantable pacemaker as example to demonstrate the abstractions and refinements of physiological models during closed-loop model checking. The results can be used for risk analysis of closed-loop medical devices.
A model-driven development toolchain automatically translates formally verified models, which represent over-approximations of the realistic models, into deterministic models which can interact with real controllers within realistic environments. The model translation process guarantees that the properties verified in the early stage were still satisfied, as the system model was refined. As the verified model is translated into executable code for physical implementation, it is validated using conformance testing procedures based on the initial system specification.
The ultimate closed-loop validation of the medical devices is clinical trial, in which the devices are directly used on patients. However, clinical trials are expensive both in time and money, and poorly-planned clinical trials lead to failure, which pose significant risks on the patients involved. In this project, we propose model-based clinical trials (MBCT), in which the devices are evaluated on a virtual patient population consists of physiological models for different patient conditions. The results of these MBCTs can be used to provide insights that can be beneficial to the actual clinical trials.
2015 © Zhihao Jiang. ALL Rights Reserved.