## Problem Overview

We propose to use Dreem headband data to perform sleep stage scoring on 30 seconds epochs of physiological signals.
Context

Sleep plays a vital role in an individual’s health and well-being. Sleep progresses in cycles that involve multiple sleep stages: wake, light sleep, deep sleep, and REM sleep. Different sleep stages are associated to di different physiological functions. Monitoring sleep stages is beneficial for diagnosing sleep disorders. The gold standard to monitor sleep stages relies on a polysomnography study conducted in a hospital or a sleep lab. Different physiological signals are recorded such as electroencephalogram, electrocardiogram  etc.

Sleep stage scoring is then performed visually by an expert on epochs of 30 seconds of signals recording. The resulting graph is called a hypnogram. It provides a compact description the night. 

![Hypnograms](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7751170/bin/zsaa097_fig4.jpg)
Fig 1: Hypnograms scored by the Dreem Headband compared to hypnograms derived by a consensus of scorers

##  The Dreem headband

Dreem headband allows doing polysomnography at home signal thanks to two kinds of sensors: electroencephalogram (EEG) and accelerometer signals.

Since the Dreem headband records a lot of nights every day, we spent time developing the most accurate automatic sleep staging algorithms.

In this challenge, we provide you with labeled data. The idea is to develop an algorithm of sleep staging able to differentiate between Wake, N1, N2, N3 and REM on windows of 30 seconds of raw data. The raw data includes 5 EEG channels in frontal and occipital positions and 3 accelerometers channels (x, y and z).

![Dreem headband](https://dreem.com/static/img/pharma/labs/headband.png)
Fig 1: The Dreem Headband

You can find a lot of information on Dreem at dreem.com