Closed- and Open-loop (Slow Oscillations) sleep stimulation (auditory) or recording in full-PSG using OpenBCI for sleep.
A poster with pictures and infographics, lists'n'stuff
- Its a fun project, but with hard science behind in hardware, software and concept, usable for research.
- Different modes: e.g. Stimulation, Recording only (just PSG, no stimulation), record on SD card without real-time monitoring/stimulation
- Setup for OpenBCI sleep recording (125 or 250 Hz sampling rate, 8 or 16 channels, incl. EEG, EOG, EMG and ECG) here
- Build your own sleep lab for under a $1000 and do research-grade recording and reactivation protocols.
- Its free, its open and its closed-loop too, at least for slow waves/slow oscillations.
- Works and tested with all Firmwares for OpenBCI device: v1 v2 v3
- Use your own Electrode Montage (for 8 and 16 channel) stored in a *.tsv to change channel setup, display in the GUI and algorithm
- Targeted & Untargeted Auditory Stimuli Presentation using Configurable Stimuli Playlists (easy to create and mix).
- Reproduce all the published research in this area, e.g. Targeted Memory Reactivation (TMR, e.g Rudoy et al. 2009) during sleep, or Slow wave enhancement (Ngo et al. 2013)
- Stimulate during a specific time point of a slow oscillation, and trigger another stimuli after that.
- Stimulation is activated manually, but disengaged automatically on arousal.
- See live ERPs from stimulation
- Latencies considered in stimulation (below 10 ms signal latency, auditory delay down to 24 ms with normal hardware under 3 ms with specialized sound card, default is 93 ms)
- 100% logging what you did, all markers, Lights-off on, checklist for eye movements
- Lost samples are imputed and logged.
- works with parallel microSD card recording (250 Hz)
- Warning on SD card problems on device
- Handles all kinds of auditory stimuli, peeps, noises, voices and baaaams! (as long as in wav or mp3 format, or what sox can handle)
- Detect accurately your hearing threshold on any PC with Linux (alsa) down to the deciBel, and stimulate accurately as well with online-adjustments of volume.
- Full unfiltered BDF+ (24bit) & CSV export or pre-filtered (High-pass, can choose cut-off frequency, Butterworth, order 1 sample) EDF+(16bit) (live recording, and including all markers)
- Obfuscate your Sham/Stim condition for blinding your experiment.
- Realtime-view, scalable channels, with spindle highlighting.
- Flexible re-referencing
- Simulation of previous Recording for testing and trying things (even without an OpenBCI device)
- Default settings for stimulation in healthy school children (5-12 yrs.) and adults (5-35 yrs.)
- ... lots of other stuff
Tutorial in the making, otherwise ask the author.
software/installation/source_python
- Example Data for testing the simulation (FREDDY is 8 channel, JINGYI is 16 channel data, settings described in here
- Example Stimuli (preprocessed) for testing the open and closed loop stimulation of voice and tones. Example protocols for those stimuli see in the subfolder stimulations
- Binaries, Ubuntu 18.04 LTS x64 compiled using pyinstaller
- Proof of concept learning Chinese in sleep files for ANKI if you like to learn more chinese.
- OpenBCI Cyton board (optionally a daisy module) v1 v2 v3
- EEG electrodes
- Battery/Power pack
- Other 'stuff' to record polysomnography
- Motivation to built a little (you can do it!) ...see tutorial here
- Dual core 1.2 Ghz
- 1.5 GB RAM
- 5 GByte free disk space
- USB 2.0 port
- Audio jack (wireless audio for stimulation not recommended, due to delay) (e.g. a Rasperry Pi3, Pine64, Rock64, etc. are fine too)
- Linux
- Ubuntu 18.04 LTS (tested), recompiled also tested under Ubunutu 14.04 LTS
- Debian 9 if source code is recompiled (tested)
- (optional, but recommended) low-latency kernel preferred (will automatically installed with the deploy scripts)
- (soon) READY ON THE STICK: setup free USB stick image for bootup in custom Debian live version on any PC soon to follow.
- Lenovo Thinkpad X220, Intel Core i7-2640M, 16 GByte RAM, fast SSD, Ubuntu 18.04 LTS
- Lenovo Thinkpad X220, Intel Core i5-2520M, 16 GByte RAM, fast SSD, Ubuntu 14.04 LTS
- Dell Laptitude, Ubuntu 18.04 LTS
- Lenovo Thinkpad X220, Intel Core i7-2640M, 16 GByte RAM, fast SSD, Ubuntu 18.04 LTS
- Lenovo Thinkpad X220, Intel Core i5-2520M, 16 GByte RAM, fast SSD, Ubuntu 14.04 LTS
- Dell Laptitude, Ubuntu 18.04 LTS
- Debian stretch (9, stable)