High-Dimensional Multivariate Time Series Data Visualization Using an Open Source Framework

Abstract

Multivariate Time Series plays a major part in statistics, signal processing, pattern recognition, econometrics, weather forecasting and earthquake prediction etc. It is very tedious task to select the appropriate technique to visualize the high-dimensional data in order to get insight or meaningful information. In this paper, we used an Open Source package named as Visbrain to visualize high-dimensional multivariate healthcare data in optimized way. It involves two dimensions of reflection: (1) questions that speak to exceedingly configurable visual natives (availability of EEG and ECG sources, and so forth.) and (2) graphical UIs for more elevated amount connections. The article level offers adaptable and measured devices for delivering and computerizing the generation of numbers with a comparative way to deal with that of Matplotlib with subplots. The second dimension outwardly associates these articles by controlling properties and communications by means of graphical interfaces. The present form of Visbrain (rendition 0.4.4) contains 14 distinct items and three custom graphical interfaces, worked with PyQt: Signal, for the control of fleeting and ghostly properties. Every module has been created in close cooperation with end clients, mostly neuroscientists and area specialists, who utilize their experience to make Visbrain as straightforward as feasible for the chronicle modalities (eg intracranial EEG, ECG, scalp EEG, MEG, anatomical and useful MRI). The paper disscuses the features and various modules for current understanding research trends in the field of Visual Analytics.