Back

Noninvasive Sleep Scoring in Mice using Electric Field Sensors

Kloefkorn, H.; Aiani, L. M.; Lakhani, A.; Nagesh, S.; Moss, A.; Goolsby, W.; Rehg, J.; Pedersen, N. P.; Hochman, S.

2019-10-07 neuroscience
10.1101/794552 bioRxiv
Show abstract

BackgroundRodent sleep scoring in principally reliant on electroencephalogram (EEG) and electromyogram (EMG), but this approach is invasive, can be expensive, and requires expertise and specialized equipment. Affordable, simple to use, and noninvasive ways to accurately quantify rodent sleep are needed.\n\nNew methodWe developed and validated a new method for sleep-wake staging in mice using cost-effective, noninvasive electric field (EF) sensors that detect respiration and other movements. We validated recordings from EF sensors attached to the exterior of specialty chambers used to continuously capture sleep with EEG/EMG, then compared this to EF sensors attached to vivarium home-cages.\n\nResultsEF sensors quantified 3-state sleep architecture (wake, rapid eye movement - REM - sleep, and non-REM sleep) with high agreement (>93%) and comparable inter- and intra-scorer error as expert EEG/EMG scoring. Novices given an instruction document with examples were able to score sleep comparable to expert scorers (>91% agreement). Additionally, EF sensors were able to quantify 3-state sleep scoring in traditional mouse home cages.\n\nComparison with existing methodMost noninvasive sleep assessment technology requires animal contact, altered cage environments, and/or can only discern 2 states of arousal (wake or asleep). The EF sensors are able to discriminate REM from non-REM sleep accurately and from outside the animals home cage.\n\nConclusionsEF sensors provide a simple and reliable method to accurately score 3-state sleep architecture; (i) from outside the typical home cage, (ii) where noninvasive approaches are preferred, or (iii) which EEG/EMG is not possible.\n\nGraphical Abstract\n\nO_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=110 SRC=\"FIGDIR/small/794552v1_ufig1.gif\" ALT=\"Figure 1\">\nView larger version (43K):\norg.highwire.dtl.DTLVardef@1b36cf3org.highwire.dtl.DTLVardef@b5fdc9org.highwire.dtl.DTLVardef@28a6dforg.highwire.dtl.DTLVardef@e39a5a_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.