Clinical Business Intelligence to Prevent Stroke Accidents

Clinical Business Intelligence to Prevent Stroke Accidents

Nuno Gonçalves (University of Minho, Portugal), Cesar Quintas (Centro Hospitalar do Porto, Portugal) and José Machado (University of Minho, Portugal)
DOI: 10.4018/978-1-4666-9882-6.ch012
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Stroke is considered the third main cause of death among all population, without distinguishing genders, led by heart diseases in first place. In other hand, despite representing a significant number of mortality, these diseases are the causes for a long-term disability in all countries with a vast recovery time going parallel with its costs. However, leaving aside this facts, stokes and heath diseases can also be easily prevented considering the outcome. This paper presents a new methodology to prevent these events to happen by using segmentation methods, which allows distinguishing and aggregating clusters of historical records, classification methods, such as Artificial Neural Networks, capable of classifying a new record according to its distribution among the clusters. A Multi-Agent Case Based Reasoning system is also proposed to evaluate solutions based in a similar case.
Chapter Preview
Top

3. Knowledge Representation

In this approach we represent data in two different ways, understandable to the user and normalized increasing classification performance. This views allows perception at different levels improving the retrieval of significant cases to be applied as relevant, or not, to a case study. There are a few main aspects that must be drilled in other to understand attributes influence and their importance in this approach.

Complete Chapter List

Search this Book:
Reset