A framework for supporting the development of verifiably safe medical best practice guideline systems

Abstract

Abstract Improving safety of patient care is an ultimate objective for medical systems. Though many medical best practice guidelines exist and are in hospital handbooks, they are often lengthy and difficult for medical professionals to remember and apply clinically. Hence, developing safe medical best practice guideline systems is an urgent need. The paper presents a framework to support the development of verifiably safe medical best practice guideline systems. The framework facilitates medical professionals’ participation in computer modeling, clinical validation, formal verification and root cause identification of safety failures at both model and code levels. To implement the framework, our strategies are to maximally utilize existing models/tools designed for validation and verification respectively, but build bridges among different selected models/tools. In particular, we use statechart tool to build statechart models for medical best practice guidelines and use statechart models to interact with medical professionals for clinical validations. The statechart models are then automatically transformed to verifiable models by the framework so that the safety properties can be formally verified. The computer models that are both validated by medical professionals and verified by formal verification tools are then used to generate computer executable code. To improve code level safety, the framework further transforms safety properties specified at the model level to runtime code monitors to ensure that these safety properties are complied at runtime. We use a simplified version of cardiac arrest treatment scenario provided to our team by Carle Foundation Hospital as a case study to evaluate the framework in developing a verifiably safe medical system.