TIME SERIES PREDICTION FOR BIOMEDICAL MEASUREMENTS USING FUZZY LOGIC

Claudio De Capua, Emilia Romeo

Abstract

In this paper is proposed an algorithm of prediction fuzzy for chaotic time series. This approach has been select because, in presence of specific pathologies, biomedical data may be represented as a chaotic time series. In particular, we are interested in monitoring the intracranial pressure (IP) of some patients in a state of coma who were suffering from intracranial hypertension syndrome. In these particular cases, prediction is necessary (from a diagnostic point of view) if you want to operate at the right moment on IP abnormal conditions. The proposed approach is based on a prediction multi-factor algorithm which doesn’t need the knowledge of the mathematical working model of the biologic phenomenon, translating the real time series into a fuzzy time series.

Download

IMEKO-TC4-2004-079.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings