Quality Checking of Medical Guidelines using Interval Temporal Logics (Preliminary report)

Computer-based decision support in health-care is becoming increasingly important in recent years. Clinical Practise Guidelines (CPGs) are documents supporting health-care professionals in managing treatment of patients in order to avoid non-standard practices or outcomes. Following ideas from publications of Hommersom, Lucas, and Balser, here we promote the use of temporal logics in order for formal specification and verification of CPGs. Unlike the works mentioned above, here we advocate the use of interval-based temporal logics instead of point-based ones. In particular, we discuss the possibility of applying Propositional Neighborhood Logic (PNL) and a metric extension of it, called MPNL, to CPGs, in order to enable explicitly take into account the duration of medical events and procedures. We argue that PNL alone can subsume the expressiveness achieved by point-based logics, and that interval-based temporal logics in general provide much more detailed and precise formalization of CPGs.