DyssomniaNN: Multi-Scale Spiking Neural Modeling of Human Sleep Disorders and Perceived Restedness
Researchers introduce DyssomniaNN, a spiking model of five sleep-disorder cohorts showing sleep time poorly predicts restedness.
DyssomniaNN couples two-process sleep regulation and a hypothalamic flip-flop circuit with a recurrent corticothalamic leaky integrate-and-fire network. It simulates healthy controls, delayed-phase disorder, chronic insomnia, paradoxical insomnia, and narcolepsy type 1. Synthetic local field potentials closely match PhysioNet Sleep-EDF spectra, and a Well-Rested Index shows sleep duration does not predict restorative recovery.
- Simulates five cohorts over an identical 72-hour photoperiod.
- Synthetic spectra match Sleep-EDF with mean absolute error 0.33-0.35.
- Paradoxical insomnia has 21.15 hours sleep but 37.6% restoration.
- Narcolepsy shows 781 minutes awake after onset and 35 SOREM periods.
- Tags
- spiking-neural-networkssleep-disorderscomputational-neuroscienceeegdyssomnianninsomnianarcolepsy
- Organizations
- PhysioNet
- AI models
- DyssomniaNN
From the original article
1 min read at arxiv.orgClinical sleep medicine repeatedly documents profound divergences between objective sleep duration and subjective daytime impairment, notably exemplified by sleep state misperception in paradoxical insomnia and boundary instability in narcolepsy type 1.
By Daniel Szelogowski · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.