Computer Science has traditionally focused on the functional aspects of design, underemphasizing the human element in the success of any technology. The failure of technologies and the accidents that happen during use require the consideration of the user and the technologies as symbiotic parts of a whole systems approach to improving diagnosis and treatment. This chapter provides an overview of the history of the critiquing approach to knowledge systems that illustrates a more human-centered approach. It is an approach that, unlike traditional knowledge-based systems, aims to provide a check on human reasoning, rather than a replacement for it. The chapter will also discuss future possibilities for research, in particular the use of social networking and recommender systems, as a means to enhance the approach.
TopKnowledge-Based Systems In Medicine
In the thirty-five years or more that knowledge-based (expert) systems have been part of computer science, there have been a number of developments in the architectures used. The earliest expert systems, for example MYCIN (Shortliffe, 1976) and DENDRAL (Buchanan 1969) had three basic components; knowledge base, user interface and inference engine. There are now several new approaches that exist as alternatives to the basic architecture, for example neural networks (Zurada, 1992), and agent-based systems (Foster et al, 2006).
Within the traditional architecture there have been many changes in the back end operation of the tools. The first expert systems primarily had a database (or knowledge base) in the form of knowledge production rules and an inference engine, which given inputs of initial information would chain through the rules to produce a diagnosis. In relation to the knowledge base, production rules were complimented with the full range of knowledge representation techniques (van Harmelen, Lifschitz, & Porter, 2008). Research has also focused on improving inference engines with new approaches and developing better user interfaces (Pandey & Mishra, 2009).
It is on the change in the approach to the user interface, that we will focus in this chapter. There are a number of ways in which decisions support and knowledge-based systems can support medical practice. Wright & Sittig conducted a review and synthesis of the history of clinical decision support systems since 1959. From this they developed a four-phase model of various architectures for integrating decision support systems with clinical systems. The four phases are: standalone decision support systems, decision support integrated into clinical systems, standards for sharing clinical decision support content and service models for decision support. The authors claim this fits with the chronological history of clinical decision support, and note that the trend is towards better integrating decision support systems into clinical workflows and other clinical technology.