Assumption-based argumentation with preferences and goals for patient-centric reasoning with interacting clinical guidelines

A paramount, yet unresolved issue in personalised medicine is that of automated reasoning with clinical guidelines multimorbidity settings. This entails enabling machines to use computerised generic guideline recommendations and patient-specific information yield patient-tailored where interactions arising due multimorbidities are resolved. problem further complicated by patient management desiderata, particular the need account for patient-centric goals as well preferences various parties involved. We propose solve this interacting context a given means computational argumentation. In particular, we advance structured argumentation formalism ABA+G (short Assumption-Based Argumentation Preferences (ABA+) Goals) integrating about recommendations, interactions, patient’s state, prioritised goals. combines assumption-based goal-driven selection among outcomes. Specifically, assume defeasible applicability general goal well-being, resolve (conflicts otherwise undesirable situations) based on state patient, employ patient-centered suggest interaction-resolving, goal-importance maximising preference-adhering recommendations. well-established Transition-based Medical Recommendation model representing identifying thereof, map components question, together goals, over actions, reasoning. this, follow principles establish corresponding theoretical properties illustrate our approach realistic scenaria.