06/12/2021 - Support for scoring more than three brain states is now available with AccuSleep X.
08/11/2020 - Mac compatibility. AccuSleep should now be functional on Mac computers.
04/09/2020 - Implemented a better algorithm for removing short bouts.
11/05/2019 - EEG/EMG data are now only loaded when necessary to avoid out-of-memory errors.
10/30/2019 - The primary user interface has received a major update, and now allows all recordings from one subject to be processed simultaneously. A small bug was also fixed, the user manual was updated, and error messages should be more helpful.
AccuSleep is a set of graphical user interfaces for scoring rodent sleep using EEG and EMG recordings. If you use AccuSleep in your research, please cite our publication:
Barger, Z., Frye, C. G., Liu, D., Dan, Y., & Bouchard, K. E. (2019). Robust, automated sleep scoring by a compact neural network with distributional shift correction. PLOS ONE, 14(12), 1–18.
The data used for training and testing AccuSleep are available at https://osf.io/py5eb/
Please contact zekebarger (at) gmail (dot) com with any questions or comments about the software.
Make sure your version of MATLAB meets the specifications in the "Requirements" section below.
Click the "Clone or download" button and choose "Download ZIP".
Extract the contents of the zip file.
Add AccuSleep to your MATLAB path. You can do this in the MATLAB "Current Folder" window by right-clicking the AccuSleep folder, clicking "Add to Path" --> "Selected Folders and Subfolders", then running the command
savepathin the Command Window.
To get started, run
AccuSleep_GUI and click the "User manual" button, or run
for a full explanation of these functions and the types of input
AccuSleep_GUI provides an interface for most of the functions
in this package, but if you want to batch process recordings from multiple subjects, you can
call the required functions yourself.
- MATLAB version 2017b or later
- Statistics and Machine Learning Toolbox
- Deep Learning Toolbox
- Signal Processing Toolbox
- Image Processing Toolbox
AccuSleep_GUIA user interface for labeling sleep states, either manually or automatically
AccuSleep_viewerA user interface for manually labeling sleep states
AccuSleep_classifyAutomatically labels sleep states using a pre-trained neural network
AccuSleep_trainTrains a neural network for labeling sleep states
createCalibrationDataGenerates a file that is required for automatic sleep state labeling for a given combination of subject and recording equipment
Tips & Troubleshooting
- Make sure the required toolboxes are installed.
- Using more data for calibration will produce better results. However, labeling more than a few minutes of each state probably isn't necessary.
- If you create a calibration file using one recording, then use it to score another recording automatically, and the accuracy is low, the signals might be different between the two recordings. In this case, it's best to create a new calibration file.
- If your accuracy seems low no matter what you do, you may wish to train your own network.
- Make sure to click the 'Help' button in AccuSleep_viewer for a list of keyboard shortcuts.
- Make sure to run
doc AccuSleep_instructionsand read the documentation before using this software.
- If your recordings are very long (>48 hours) and are not displaying properly, try splitting them into smaller files.
- Ensure the epoch length associated with the labels, calibration data, and trained network are the same.
- Networks trained using MATLAB 2019a or later do not seem to be backward compatible with earlier versions of MATLAB. However, networks trained on 2018b or earlier seem to be forward compatible.
- Make sure the recordings are free of NaN and Inf values.
- Please contact zeke (at) berkeley (dot) edu if you find any other issues.
We would like to thank Franz Weber for creating an early version of the manual labeling interface.