This is a past event
Abstract:
This talk shows how two different logic-based approaches can be used to (1) detect conflicts in the treatment of multiple chronic conditions, and (2) modelling cancer and tumour growth.
(1) We developed an automated approach focusing on the composition of (static or behavioural) design models via constraint solvers. This approach is directly applicable to many different domains, and can be used for the detection of conflicts in clinical pathways. Clinical pathways are care plans which detail essential steps in the care of patients with a specific clinical problem, usually a chronic disease. A pathway includes recommendations of medications prescribed at different stages of the care plan. For patients with three or more chronic diseases (known as multimorbidities) the multiple pathways have to be applied together. One common problem for such patients is the possible adverse interaction between medications given for different diseases. In this context, we have used an optimising SMT solver (Z3) to quickly find the set of medications with the minimal number and severity of conflicts which is assumed to be the safest.
(2) Automata can be used to capture cancer evolving through a (discrete finite) sequence of progressive stages called phenotypes. Automata consist of states (known as hallmarks of cancer) and transitions between states, indicating a progression or regression of the cancer. We haved explored extensions and combinations of different variants of timed automata and associated tools to model and analyse a model of the disease in different ways. We combine patient information and comorbidities with the cancer automaton through composition. We have explored the use of model checking as an analysis technique to provide further insights into the effectiveness of treatment plans for a given patient, and how these could potentially inhibit or slow down the progression of cancer.
Bio:
Juliana Bowles (née Küster-Filipe) is a lecturer at the School of Computer Science, University of St Andrews, Scotland. She has research interests in formal methods and dependability, logics and models for true-concurrency, and foundations of modelling languages. She is currently applying her foundational work to the health care domain, and has an EPSRC-funded project on "Automated Conflict Resolution in Clinical Pathways".
Bowles received her PhD (Dr.rer.nat) from the Technical University of Braunschweig, Germany, in 2000.
- Speaker
- Juliana Bowles
- Hosted by
- Adam Wyner
- Venue
- Meston 311