Addressing the need for adaptable decision processes within healthcare software.

Abstract

In the healthcare sector, where the decisions made by software aid in the direct treatment of patients, software requires high levels of assurance to ensure the correct interpretation of the tasks it is automating. This paper argues that introducing adaptable decision processes within eHealthcare initiatives can reduce software-maintenance complexity and, due to the instantaneous, distributed deployment of decision models, allow for quicker updates of current best practice, thereby improving patient care. The paper provides a description of a collection of technologies and tools that can be used to provide the required adaptation in a decision process. These tools are evaluated against two case studies that individually highlight different requirements in eHealthcare: a breast-cancer decision-support system, in partnership with several of the UK's leading cancer hospitals, and a dental triage in partnership with the Royal Liverpool Hospital which both show how the complete process flow of software can be abstracted and adapted, and the benefits that arise as a result.