A general framework for time series data mining based on event analysis: Application to the medical domains of electroencephalography and stabilometry

Abstract

There are now domains where information is recorded over a period of time, leading to sequences of data known as time series. In many domains, like medicine, time series analysis requires to focus on certain regions of interest, known as events, rather than analyzing the whole time series. In this paper, we propose a framework for knowledge discovery in both one-dimensional and multidimensional time series containing events. We show how our approach can be used to classify medical time series by means of a process that identifies events in time series, generates time series reference models of representative events and compares two time series by analyzing the events they have in common. We have applied our framework on time series generated in the areas of electroencephalography (EEG) and stabilometry. Framework performance was evaluated in terms of classification accuracy, and the results confirmed that the proposed schema has potential for classifying EEG and stabilometric signals. The proposed framework is useful for discovering knowledge from medical time series containing events, such as stabilometric and electroencephalographic time series. These results would be equally applicable to other medical domains generating iconographic time series, such as, for example, electrocardiography (ECG).

Beyond spikes: Multiscale computational analysis ofin vivolong-term recordings in the cockroach circadian clock

“…In order to find groups of similar events, and estimate the set {Θ R ( t ), Θ 1 ( t ), Θ 2 ( t ), …, Θ K ( t )} posed in Equation 2, clustering was performed using estimated widths, amplitudes, and time of each event as features. This election of features was also proposed in Lara et al (2014) when performing data mining of EEG recordings. These features were first rescaled, in order to make them comparable in the calculation of Euclidean distances.…”

“…These observations served to hint at the structural configuration of the system and will be helpful for further analysis of connectivity of the cockroach circadian clock’s neural network. To infer the possible interrelations of events, the events were grouped additionally by similarities through a clustering process (Lara, Lizcano, Pérez, & Valente, 2014) (Figure S1). Bayesian Gaussian mixtures were applied, using the estimated duration, amplitude, and time localization as coordinates.…”

Clinical Information Systems and Artificial Intelligence: Recent Research Trends

“…The literature includes some papers on specific application domains where the use of LDAPPM seems to lead to promising results. The paper by Lara et al 16 in the Journal of Biomedical Informatics aims at deploying data mining techniques on data related to events in time series. The paper focuses on EEG (electroencephalography) data, and it uses data from a publically available data source (EEG recordings), while extension of the approach to data from clinical information systems is straightforward.…”

A method for outlier detection based on cluster analysis and visual expert criteria

“…Our proposal is evaluated according to a similarity matrix between each pair of time series of each individual. To build the above matrix, we used a time series comparison method reported elsewhere (Lara, ; Lara, Lizcano, Pérez, & Valente, ) and explained in Section 4.1. The data used to validate our proposal in the stabilometry and EEG domains are described in Sections 4.2 and 4.3, respectively.…”

“…If the series do not have any event in common, similarity will be equal to 0. The method used for this purpose was originally proposed elsewhere (Lara et al, ). It is outlined here less formally, using the original notation.…”