Improving Medical Safety using Software Engineering Technology

This project will explore the application of software engineering technology to the improvement of medical care. Medical care is delivered by doctors, nurses, hospital staff, etc., by performing processes intended to coordinate the efforts of all, and to make efficient use of scarce resources such as expensive equipment and doctors' time and attention. These processes are surprisingly large and complex, and it is not unusual for them to be performed incorrectly. In the US nearly 100,000 deaths a year are attributable to medical errors, and many of them appear to be due to process misunderstandings, errors in performance of process tasks, or defects in the processes themselves. Software engineering research has, for many years, understood the importance of process, and has demonstrated how technologies can improve outcomes by improving processes in the domain of software development. Preliminary research by visiting researchers Osterweil and Clarke in the US has demonstrated that software technologies can bring improvements of these kinds in the domain of medical practice as well. Additional design technologies developed by PI Rosenblum and collaborator Kramer should bring additional types of improvements. The purpose of this project is to explore that premise.The project will bring together researchers from the UK and the USA whose complementary strengths will bring new insights, that should lead to new technologies, and safer medical care. Osterweil's work on process definition will support the precise and rigorous definition of medical processes, and of the resources used in performing them. Kramer's work on viewpoints and requirements will help the researchers communicate with medical professionals better, thereby resulting in clearer and more precise understandings of the processes to be defined. Clarke's work on software analysis will formalise the safety characteristics that are required, and will apply analysers to identify (potentially lethal) defects in the processes. Rosenblum's work will provide simulations that the researchers will use to evaluate possible efficiencies obtained from process improvements.The research will result in medical processes that are demonstrably superior, and thus safer. Grappling with processes in this new domain is also expected to reveal inadequacies on the four researchers' existing technologies, thereby also leading to improvements in the technologies as well.