The DREAMS Artifacts Database

Contact:
Stéphanie DEVUYST
Université de Mons
Faculté Polytechnique de Mons - TCTS Lab
31, Boulevard Dolez
B-7000 Mons (Belgium)
ph : +32 (65) 37.47.20
fax: +32 (65) 37.47.29
stephanie.devuyst@umons.ac.be


Introduction Description Format Download More

Introduction

This artifacts database was collected during the DREAMS project, to tune, train and test our automatic detection algorithms.

It consist of 20 excerpts of 15 minutes of PSG recordings (extracted from whole-nights), annotated by an expert in artifacts.

In order to facilitate future research and performance comparision, we decided to publish these data on Internet. Be simply aware that this DREAMS Artifacts Database of University of MONS - TCTS Laboratory (Stéphanie Devuyst, Thierry Dutoit) and Université Libre de Bruxelles - CHU de Charleroi Sleep Laboratory (Myriam Kerkhofs) is under terms of the Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0) License. Each use of this database and/or its content must be attributed to University of MONS - TCTS Laboratory (Stéphanie Devuyst, Thierry Dutoit) and Université Libre de Bruxelles - CHU de Charleroi Sleep Laboratory (Myriam Kerkhofs). Any other additional authorisations may be asked to avre@umons.ac.be.

If you are interested in adding your own visual or automatic annotations, please contact me: stephanie.devuyst@umons.ac.be


Database description

Data collected were acquired in a sleep laboratory of a belgium hospital using a digital 32-channel polygraph (BrainnetTM System of MEDATEC, Brussels, Belgium). They consist of 20 polysomnographic recordings coming from patients with different pathologies (dysomnia, restless legs syndrome, insomnia, apnoea/hypopnoea syndrome). Two EOG channels (P8-A1, P18-A1), three EEG channels (CZ-A1 or C3-A1, FP1-A1 and O1-A1) and one submental EMG channel were recorded. The standard European Data Format (EDF) was used for storing. The sampling frequency was 200Hz, 100Hz or 50Hz. An excerpt of 15 minutes was extracted from each of these recordings to be annotated in artifacts by an expert. These excerpts were voluntarily withdrawn at various times of the night, in order to diversify the background of the physiological signals observed (depending on sleep stage considered). The corresponding whole-night recordings were also previously scored in sleep stages according to the Rechtschaffen and Kales criteria, but no note of these stages quotations was given to the scorers.

Name Sampling frequency Age Sex Time of the excerpt
(from the begining of the recording)
excerpt1 50Hz 60 Woman extracted from 00:20:00 to 00:35:00
excerpt2 50Hz 53 Man extracted from 01:23:00 to 01:38:00
excerpt3 100Hz 50 Woman extracted from 08:17:00 to 08:32:00
excerpt4 100Hz 51 Man extracted from 00:14:00 to 00:29:00
excerpt5 200Hz 73 Man extracted from 05:14:00 to 05:29:00
excerpt6 100Hz 42 Man extracted from 03:45:00 to 04:00:00
excerpt7 50Hz 65 Man extracted from 04:03:00 to 04:18:00
excerpt8 200Hz 53 Man extracted from 07:38:00 to 07:53:00
excerpt9 50Hz 63 Man extracted from 00:46:00 to 01:01:00
excerpt10 200Hz 72 Man extracted from 03:15:00 to 03:30:00
excerpt11 200Hz 47 Man extracted from 04:20:00 to 04:35:00
excerpt12 200Hz 31 Man extracted from 00:17:00 to 00:32:00
excerpt13 200Hz 52 Woman extracted from 01:50:00 to 02:05:00
excerpt14 200Hz 42 Man extracted from 08:01:00 to 08:16:00
excerpt15 200Hz 40 Man extracted from 08:04:00 to 08:19:00
excerpt16 200Hz 52 Man extracted from 02:22:00 to 02:37:00
excerpt17 200Hz 53 Woman extracted from 01:49:00 to 02:04:00
excerpt18 200Hz 61 Woman extracted from 00:00:00 to 00:15:00
excerpt19 200Hz 73 Man extracted from 05:39:00 to 05:54:00
excerpt20 200Hz 48 Man extracted from 03:00:00 to 03:15:00

The artifacts asked to be identified in these excerpts were the following ones (click on link to see example):

A summary of the recordings contents in artifacts is presented below. Unfortunately, no saturation has been reported by the expert in this database. We also note that only one period of EEG was annotated as corrupted by a network interference. However, this is hardly surprising given that the Medatec acquisition system already includes filters to eliminate these spurious frequencies.

Atf_ondE
Atf_ondO
Atf_cardE
Atf_cardO
Atf_50HzE
Atf_50HzO
Atf_50HzM
Atf_satE
Atf_satO
Atf_satM
Atf_highE
Atf_noE
Atf_noO
Atf_tansE
Atf_mvtE
Atf_phaseO
Atf_otherE
Atf_otherO
excerpt1 X X X X X X X
excerpt2 X X X X X
excerpt3 X X X X X X X X
excerpt4 X X X X X X
excerpt5 X X X X X
excerpt6 X X X X X X X
excerpt7 X X X X X X X X
excerpt8 X X X X X X X X
excerpt9 X X X X X
excerpt10 X X X
excerpt11 X X X X X
excerpt12 X X X X
excerpt13 X X X X X
excerpt14 X X X X X X
excerpt15 X X X X X X X
excerpt16 X X X X X X
excerpt17 X X X X
excerpt18 X X X X
excerpt19 X X X X X
excerpt20 X X X X X X

Return to menu


File format

Four files are available per excerpt:

Return to menu


Database download

You can download each file separately by doing "right-click" and "save target as" on the links below (in the table),
or you can directly download the entire database by clicking on this link: DatabaseArtifacts.rar

EDF Automatic detection Visual scoring 1 Hypnogram
excerpt1.edf Automatic_detection_excerpt1.txt Visual_scoring1_excerpt1.txt Hypnogram_excerpt1.txt
excerpt2.edf Automatic_detection_excerpt2.txt Visual_scoring1_excerpt2.txt Hypnogram_excerpt2.txt
excerpt3.edf Automatic_detection_excerpt3.txt Visual_scoring1_excerpt3.txt Hypnogram_excerpt3.txt
excerpt4.edf Automatic_detection_excerpt4.txt Visual_scoring1_excerpt4.txt Hypnogram_excerpt4.txt
excerpt5.edf Automatic_detection_excerpt5.txt Visual_scoring1_excerpt5.txt Hypnogram_excerpt5.txt
excerpt6.edf Automatic_detection_excerpt6.txt Visual_scoring1_excerpt6.txt Hypnogram_excerpt6.txt
excerpt7.edf Automatic_detection_excerpt7.txt Visual_scoring1_excerpt7.txt Hypnogram_excerpt7.txt
excerpt8.edf Automatic_detection_excerpt8.txt Visual_scoring1_excerpt8.txt Hypnogram_excerpt8.txt
excerpt9.edf Automatic_detection_excerpt9.txt Visual_scoring1_excerpt9.txt Hypnogram_excerpt9.txt
excerpt10.edf Automatic_detection_excerpt10.txt Visual_scoring1_excerpt10.txt Hypnogram_excerpt10.txt
excerpt11.edf Automatic_detection_excerpt11.txt Visual_scoring1_excerpt11.txt Hypnogram_excerpt11.txt
excerpt12.edf Automatic_detection_excerpt12.txt Visual_scoring1_excerpt12.txt Hypnogram_excerpt12.txt
excerpt13.edf Automatic_detection_excerpt13.txt Visual_scoring1_excerpt13.txt Hypnogram_excerpt13.txt
excerpt14.edf Automatic_detection_excerpt14.txt Visual_scoring1_excerpt14.txt Hypnogram_excerpt14.txt
excerpt15.edf Automatic_detection_excerpt15.txt Visual_scoring1_excerpt15.txt Hypnogram_excerpt15.txt
excerpt16.edf Automatic_detection_excerpt16.txt Visual_scoring1_excerpt16.txt Hypnogram_excerpt16.txt
excerpt17.edf Automatic_detection_excerpt17.txt Visual_scoring1_excerpt17.txt Hypnogram_excerpt17.txt
excerpt18.edf Automatic_detection_excerpt18.txt Visual_scoring1_excerpt18.txt Hypnogram_excerpt18.txt
excerpt19.edf Automatic_detection_excerpt19.txt Visual_scoring1_excerpt19.txt Hypnogram_excerpt19.txt
excerpt20.edf Automatic_detection_excerpt20.txt Visual_scoring1_excerpt20.txt Hypnogram_excerpt20.txt

Return to menu


More

Return to menu