Information Technology Reference
In-Depth Information
Bayesian Network-Based Model for the Diagnosis
of Deterioration of Semantic Content Compatible
with Alzheimer's Disease
José María Guerrero Triviño, Rafael Martínez-Tomás
and Herminia Peraita Adrados
josemaria.guerrero@cpiia.org
Abstract. Alzheimer's Disease (AD) has become a serious public health
problem that affects both the patient and his family and social environ-
ment, not to mention the high economic cost for families and public ad-
ministrations. The early detection of AD has become one of the principal
focuses of research, and its diagnosis is fundamental when the disease is
incipient or even prodromic, because it is at these stages when treatments
are more effective. There are numerous research studies to characterise
the disease in these stages, and we have used the specific research car-
ried out by Drs. Herminia Peraita and Lina Grasso. The application of
Artificial Intelligence techniques, such as Bayesian Networks and Influ-
ence Diagrams, may provide a very valuable contribution both to the
very research and the application of results. This article justifies using
Bayesian Networks and Influence Diagrams to solve this type of prob-
lems and because of their great contribution to this application field.
The modelling techniques used for constructing the Bayesian Network
are mentioned in this article, and a mechanism for automatic learning of
the model parameters is established.
Keywords: Bayesian Network, Influence Diagram, Corpus of Oral Def-
initions, Naive Bayes, Alzheimer's Disease, Cognitive Deterioration.
1
Introduction
As in other fields in the real world, medical diagnosis is not always 100% accu-
rate. In the specific case of Dementia and especially Alzheimer's Disease (AD), its
diagnosis is sometimes an extremely dicult task, especially when it is incipient
and intensity is only slight[15]. These are the main reasons for using Soft Com-
puting techniques (techniques that enable us to work with incomplete, inexact
and uncertain information) to solve this type of problems. Bayesian Networks,
in particular, provide a probabilistic model that makes it possible to define the
causal relations between the variables explicitly. This causality is assigned a rela-
tion force, which is logically determined by the degree of correlation or causality
between the variables. Bayesian Networks are extremely useful in response to
new cases,and there are Automatic-Learning techniques for both qualitative and
quantitative models. These Automatic Learning techniques can enable us to
 
Search WWH ::




Custom Search