Alzheimer detection model applying artificial intelligence techniques

Abstract

For several years now, several computational processes have been investigated in conjunction with the new techniques of artificial intelligence, data mining and the use of algorithms that are in line with the classification of data that they have contributed in other contexts such as The case in the areas of medicine especially in neurological diseases such as Alzheimer's, Parkinson's and Huntington's for this case was the study of Alzheimer's disease that affects the elderly in their cognitive and behavioral functions, the objective of this research is oriented to manipulate existing data of the disease supported by the use of the negative selection algorithm that is part of a classification of the context of genetic algorithms, classification techniques use decision trees, Naive Bayes, System Networks Neuronal Supervised Classifier and, the decision trees represent the decisions of the In the data, naive bayes is exclusive of a probabilistic context, the neural network that is based on an environment of the biological model, UCS is identified in a type of rules of a condition and action. The results have been from the origin of data from a patient population of 499 original instances and 1001 generated by means of the negative selection algorithm, the detection rate and the false alarm rate after the mathematical process justifies the validity of the data evidencing the way in which the data have been properly manipulated and verified, finally it is concluded that this type of algorithms is the case of negative selection and also in applications with conditions similar to neurological contexts as in the case of this investigation. It can also be stated that the classification techniques and their applicability make it possible to determine the use of metaclassifiers that are subject to the identification and verification of the results of the research carried out. © 2018 AISTI.