DETECT: Detection of Events in Continuous Time Toolbox: Users Guide, Examples, and Function Reference Documentation

Report No. ARL-SR-269
Authors: Vernon Lawhern; W. David Hairston; Kay Robbins
Date/Pages: June 2013; 56 pages
Abstract: DETECT (Detection of Events in Continuous Time) is a MATLAB toolbox for automated event detection in long, continuous multichannel time series. Although developed for electroencephalography (EEG), it uses a universal format that is applicable to many types of physiological time-series data or case uses benefitting from rapid, automated discrimination of specific predefined signals, and is free-standing (requiring no other plugins or packages). The primary goal is a toolbox that is simple for researchers to use, allowing them to quickly train a model on multiple classes of events, assess the accuracy of the model, and determine how closely the results agree with their own manual identification of events without requiring extensive programming knowledge or machine learning experience. Here, we provide reference documentation covering use of the DETECT toolbox, including an overview, explanations of each of the primary components and how they interact, and full help documentation for each function in the toolbox. Additionally we provide six example uses of the toolbox, including labeling trials, labeling continuous time series, manually labeling data, plotting labeled data, updating previously labeled dataset, and comparing two labeled datasets.
Distribution: Approved for public release
  Download Report ( 0.919 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.

Last Update / Reviewed: June 1, 2013