Non-invasive Classifier for Sleep and Wake Behaviors in Mice
Author Information
Author(s): Kevin D Donohue, Dharshan C Mendoza, Eli R Crane, Bruce F O'Hara
Primary Institution: University of Kentucky
Hypothesis
Can a non-invasive high-throughput system effectively classify sleep and wake behaviors in mice?
Conclusion
The developed system achieved a classification rate of 94% for sleep and wake behaviors in mice, demonstrating its effectiveness.
Supporting Evidence
- The system achieved a classification rate of 94% compared to human observations.
- Data was collected over 28.5 hours from 4 mice.
- Classification performance improved with longer segment lengths and amplitude compression.
Takeaway
Researchers created a system that can tell if mice are awake or asleep just by sensing their movements, making it easier to study their sleep patterns.
Methodology
The study used a PVDF sensor to detect pressure changes on the cage floor, classifying behaviors based on motion signals.
Potential Biases
Human observation may introduce bias in classifying sleep states.
Limitations
The system does not assess actual sleep states relative to EEG measurements and may misclassify some behaviors.
Participant Demographics
Four C57BL/6J mice, 2 males and 2 females, approximately six months old.
Statistical Information
P-Value
0.0001
Confidence Interval
95%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website