Natural Intelligence Based Automatic Knowledge Discovery for Medical Practitioners

Abstract

Medical data in the form of electronic patient records has grown manifold over the past few years. This data contains a plethora of hidden information which could prove extremely useful to the medical practitioner and contribute to the advancement of medical science in general. This chapter focuses on techniques that can be used to reveal interesting information in the form of associations between attributes in the medical domain. These techniques have been derived from nature and can be applied successfully to improve the performance of the rule mining process. Specifically, the rule mining algorithms are based on concepts of swarm intelligence and natural behavior of frogs. Swarm Intelligence (SI) is the property of a system whereby the collective behavior of agents interacting locally with their environment causes coherent functional global patterns to emerge. SI provides the foundation for exploring collective (or distributed) problem solving without centralized control or the provision of a global model. Association rule mining aims to extract interesting correlations, frequent patterns, associations or casual structures among sets of items in data repositories. Rules are a prime formalism for expressing knowledge in a symbolic way. Rules have advantages of simplicity, uniformity, transparency, and ease of inference which makes them a suitable approach for representing real world medical knowledge. This chapter discusses two new algorithms for rule mining, their implementation and analysis of performance over a medical database. Results indicate that the usability of the rules thus uncovered, is high in the medical domain, and it can be further improved by refining the fitness function. The rule mining techniques may be used to construct an automatic medical knowledge discovery system which can aid in diagnosis, prognosis, monitoring and treatment planning. Section 5.1 discusses the key concepts of rule mining and swarm intelligence. Section 5.2 describes conventional rule mining techniques and states the rationale behind using swarm intelligence and frog leaping for rule mining and classification. Section 5.3 elaborates the various algorithms that have been implemented in our study. Section 5.4 describes the details of the experiment. Section 5.5 presents the results of the practical experiment followed by conclusions and scope for further research in Sect. 5.6.