Thursday, June 11, 2015

Recording sleepwalking

Recording sleepwalking

In order to understand my sleepwalking I needed first to be able to exhaustively record it. In other words to use an approach that identify every instance of sleepwalking.

Video monitoring to track sleepwalking

Since video monitoring is an essential part of sleepwalking studies at sleep clinics and since sleepwalking is an obvious behavior (wandering when supposed to be sleeping) I recorded my nights using a surveillance video monitoring kit.
The key features being:
  • continuous recording
  • night vision (infrared based)
  • time stamp (to the second)

For each night I can then identify all sleepwalking events with the following characteristics: 
  • time of event start
  • length of event
  • event description

EEG data

At this point I am not recording any EEG data, and explain why here.

Event tracking using sleep tracking devices and apps

In complement to video I am also recording data for each night event using multiple sleep tracking devices, to add the following information:
  • start time of sleep
  • heart rate
  • sleep stage at time of event (see discussion of the validity of those data)
  • testing of sleep tracking devices capabilities for sleepwalking monitoring (see sleep trackers comparison)

Using those 2 approaches I can then characterize each event, which open the way to:
  • obtain a "sleepwalking" score for each night
  • compare events
  • try to find correlations

What does my sleepwalking look like?

Here are the data.


  1. You may describe somewhere in the posts, but is there some reason you are not using the Zeo hardware and programs to identify sleep. I have found it helpful in identifying REM, since my issue is REM sleep Behavior Disorder (RBD). Mostly my behavior is thrashing out and talking or yelling when in REM, but occasionally involves out-of-bed mobility.

    What surveillance video monitoring equipment are you using - I need to SEE what is going on. Currently I am audio recording my sleep and viewing it with Audacity.

    What software do you use to quickly scan through hours of sleep?

  2. Hi Robert,

    I haven't used Zeo and I am not sure I can still get one since the company making them went down a few years ago. But I am considering using axbo, although it is probably not as useful since they use a wristband instead of a headband for Zeo.

    As for video monitoring I am using a regular home surveillance camera system since it continuously records (not just when there is movement) and I can easily get access to all the video files. I added a strong IR light next to the camera as the built-in IR light was too weak to really illuminate the bed area.
    As for reading the video files I use the built-in software that came with the surveillance system as it offers a 16x fast forward option, making that I need only 20-30 minutes to go through a full night. Most of the video is about me not moving so it is pretty quick to go through that. The longer part is when I do sleepwalk as I record a bench of parameters about the event.

    Thanks for pointing out the Zeo device and I hope you get to elucidate your RBD soon.