BioSig: The Free and Open Source Software Library for Biomedical Signal Processing
Abstract
BioSig is an open source software library for biomedical signal processing. The aim of the BioSig project is to foster research in biomedical signal processing by providing free and open source software tools for many different application areas. Some of the areas where BioSig can be employed are neuroinformatics, brain-computer interfaces, neurophysiology, psychology, cardiovascular systems, and sleep research. Moreover, the analysis of biosignals such as the electroencephalogram (EEG), electrocorticogram (ECoG), electrocardiogram (ECG), electrooculogram (EOG), electromyogram (EMG), or respiration signals is a very relevant element of the BioSig project. Specifically, BioSig provides solutions for data acquisition, artifact processing, quality control, feature extraction, classification, modeling, and data visualization, to name a few. In this paper, we highlight several methods to help students and researchers to work more efficiently with biomedical signals.
Introduction
The area of biomedical signal processing has to deal with a large variety of topics. Artifact contamination, low signal-to-noise ratios, different data formats, classification, and statistical evaluation are general challenges of the field. Furthermore, a large number of different data processing methods for different signal modalities (EEG, ECG, etc.) and for different applications has to be considered. Moreover, software development itself is an important part of biomedical signal processing.
In the 1990s, the use of Matlab became popular to process biosignals. However, the algorithms were rarely available and the reimplementation of methods was common, even within the same research group. The field of software development was characterized by providers that offered closed (proprietary) solutions. This caused incompatibilities, and the same algorithms were implemented again and again. Another side effect was that each equipment provider defined its own data format for storing biosignals. These data could then be analysed only with the proprietary software of the vendor. Data export, if possible, was difficult and resulted usually in loss of information (e.g., metadata about the recording conditions, like filter settings or sampling rate, were not preserved).
These facts made the development and validation of new methods difficult. Additionally, the success of free and open source software in the field of operating systems (e.g., Linux) and server software did encourage the development of a free software library for biomedical signal processing.
Despite its focus on EEG data, BioSig can be used for general signal processing tasks related to a variety of measurement modalities. One example is the calculation of event-related averages in MEG. Another one is the calculation of spectral estimates of individual channels or time segments in functional near-infrared spectroscopy data. BioSig covers many EEG and polygraphic data formats. Furthermore, data loading is accomplished by a simple command.
- (i)
BioSig for Octave and Matlab (biosig4octmat). A toolbox for Octave and Matlab with powerful data import and export filters, feature extraction algorithms, classification methods, and a powerful viewing and scoring software.
- (ii)
BioSig for C/C++ (biosig4c++). A C/C++ library that provides reading and writing routines for different biosignal data formats.
- (iii)
rtsBCI (rtsbci). A real-time Brain Computer Interface (BCI) system implemented in Matlab and Simulink.
Most functions implemented in BioSig can be used with both Matlab and Octave and are installed through the package “biosig4octmat”. This is also the main module of the project. Within this library many data formats are supported, and the toolbox provides a common interface for reading different formats. An automated detection of the file format eases the use, making the detection transparent to the user. The writing of several common file formats is also supported. Additionally, useful algorithms for artifact detection and correction are available. Many algorithms for stochastic model parameters (autoregressive, multivariate, time-varying, etc.) are accessible in the time series analysis (TSA) toolbox. These and other functions from the NaN-toolbox are able to handle data with missing values (caused by, e.g., artifacts), too.
BioSig software is available “on-line” and under the terms of the “General Public License” (GPL) v3 . The GPL guarantees to the users that the BioSig library can be freely used, studied, modified, and distributed. Having a library for biomedical signal processing provides a summary of prior art in the field and might be helpful against the detrimental effects of software patents.
No comments:
Post a Comment