An overview of online based platforms for sharing and analyzing electrophysiology data from big data perspective

Abstract

With the development of applications and high‐throughput sensor technologies in medical fields, scientists and scientific professionals are facing a big challenge—how to manage and analyze the big electrophysiological datasets created by these sensor technologies. The challenge exhibits several aspects: one is the size of the data (which is usually more than terabytes); the second is the format used to store the data (the data created are generally stored using different formats); the third is that most of these unstructured, semi‐structured, or structured datasets are still distributed over many researchers' own local computers in their laboratories, which are not open access, to become isolated data islands. Thus, how to overcome the challenge and share/mine the scientific data has become an important research topic. The aim of this paper is to systematically review recent published research on the developed web‐based electrophysiological data platforms from the perspective of cloud computing and programming frameworks. Based on this review, we suggest that a conceptual scientific workflow‐based programming framework associated with an elastic cloud computing environment running big data tools (such as Hadoop and Spark) is a good choice for facilitating effective data mining and collaboration among scientists. WIREs Data Mining Knowl Discov 2017, 7:e1206. doi: 10.1002/widm.1206 This article is categorized under: Application Areas > Health Care Fundamental Concepts of Data and Knowledge > Information Repositories Technologies > Computer Architectures for Data Mining

Parallel Statistical and Machine Learning Methods for Estimation of Physical Load

“…That is why the much larger datasets and additional research will be necessary for the more collec-tively and personally tailored models. In this context, the further progress can be reached by sharing the similar datasets around the world in the spirit of open science, volunteer data collection, processing and computing [37][38][39]. The data and results presented allow us to extend application of these methods for modeling and characterization of complex human activity patterns.…”

Prediction of Physical Load Level by Machine Learning Analysis of Heart Activity after Exercises

“…But for this purpose additional research with usage of much larger datasets will be crucially important. In this connection, the more promicing progress can be obtained by creating and sharing the similar datasets around the world in the spirit of open science, volunteer data collection, processing and computing [39]- [41].…”

Relationship between the Spectral Power Density of Sleep Electroencephalography and Psychiatric Symptoms in Patients with Breathing-related Sleep Disorder

“…This open-source Matlab graphic interface is designed for the spectral analysis of sleep EEG in PSG. It automatically detects and deletes epochs with artifacts and yields summary figures for visual adjudication [ 31 , 32 ]. According to Welch’s method, it calculates the spectral power density using ten overlapping 4-second sub-epochs for each 30-second epoch with a 50% tapered cosine window.…”