DataPDF Available

MindBigData the MNIST of Brain Digits v1.01

Authors:
David Vivancos at MindBigData.com
David Vivancos
  • MindBigData.com
Download file PDF
Read file
Download citation

Abstract

The version 1.01 of the open database contains 1.160.193 brain signals of 2 seconds each, captured with the stimulus of seeing a digit (from 0 to 9) and thinking about it, over the course of almost 2 years between 2014 & 2015, from a single Test Subject David Vivancos. All the signals have been captured using commercial EEGs (not medical grade), NeuroSky MindWave, Emotiv EPOC, Interaxon Muse & Emotiv Insight, covering a total of 19 Brain (10/20) locations. We built our own tools to capture them, but there is no post-processing on our side, so they come raw as they are read from each EEG device, in total 383,751,556 Data Points. Feel free to test any machine learning, deep learning or whatever algorithm you think it could fit, we only ask for acknowledging the source and please let us know of your performance! We choose not to differentiate the signals into training/test/validation sets at this point so pick the distribution you prefer. A small portion of the signals were captured without the stimulus of seeing the digits for contrast, all are random actions not related to thinking or seeing digits, you can decide to use them or not in your tests, they use the code -1. FILE FORMAT: The data is stored in a very simple text format including: [id]: a numeric, only for reference purposes. [event] id, a integer, used to distinguish the same event captured at different brain locations, used only by multichannel devices (all except MW). [device]: a 2 character string, to identify the device used to capture the signals, "MW" for MindWave, "EP" for Emotive Epoc, "MU" for Interaxon Muse & "IN" for Emotiv Insight. [channel]: a string, to indentify the 10/20 brain location of the signal, with possible values: MindWave "FP1" EPOC "AF3, "F7", "F3", "FC5", "T7", "P7", "O1", "O2", "P8", "T8", "FC6", "F4", "F8", "AF4" Muse "TP9,"FP1","FP2", "TP10" Insight "AF3,"AF4","T7","T8","PZ" [code]: a integer, to indentify the digit been thought/seen, with possible values 0,1,2,3,4,5,6,7,8,9 or -1 for random captured signals not related to any of the digits. [size]: a integer, to identify the size in number of values captured in the 2 seconds of this signal, since the Hz of each device varies, in "theory" the value is close to 512Hz for MW, 128Hz for EP, 220Hz for MU & 128Hz for IN, for each of the 2 seconds. [data]: a coma separated set of numbers, with the time-series amplitude of the signal, each device uses a different precision to identify the electrical potential captured from the brain: integers in the case of MW & MU or real numbers in the case of EP & IN. There is no headers in the files, every line is a signal, and the fields are separated by a tab For example one line of each device could be (without the headers) [id] [event] [device] [channel] [code] [size] [data] 27 27 MW FP1 5 952 18,12,13,12,5,3,11,23,37,36,26,24,35,42…… 67650 67636 EP F7 7 260 4482.564102,4477.435897,4484.102564……. 978210 132693 MU TP10 1 476 506,508,509,501,497,494,497,490,490,493…… 1142043 173652 IN AF3 0 256 4259.487179,4237.948717,4247.179487,4242.051282…… Contact us if you need any more info. Let's decode My Brain! September 2015 David Vivancos vivancos@vivancos.com
Content uploaded by David Vivancos
Author content
All content in this area was uploaded by David Vivancos on Sep 16, 2015
Content may be subject to copyright.
16/9/2015 MindBigDatatheMNISTofBrainDigits
http://www.mindbigdata.com/opendb/index.html 1/2
MindBigData
The"MNIST"ofBrainDigits
Theversion1.01oftheopendatabasecontains1.160.193brainsignalsof2secondseach,capturedwiththestimulusof
seeingadigit(from0to9)andthinkingaboutit,overthecourseofalmost2yearsbetween2014&2015,fromasingle
TestSubjectDavidVivancos.
AllthesignalshavebeencapturedusingcommercialEEGs(notmedicalgrade),NeuroSkyMindWave,EmotivEPOC,
InteraxonMuse&EmotivInsight,coveringatotalof19Brain(10/20)locations.
Fourfilesareavailablefordownload:
DataBase File Zipsize UncompressedFilesize
MindWave MindBigDataMWv1.0.zip 62,6MB(65,663,303bytes) 297MB(311,994,495bytes)
EPOC MindBigDataEPv1.0.zip 409MB(429,732,466bytes) 2,66GB(2,859,712,035bytes)
Muse MindBigDataMUv1.0.zip 62,6MB(65,663,303bytes) 297MB(311,994,495bytes)
Insight* MindBigDataINv1.01.zip 8,49MB(8.903.725bytes) 53,2MB(55.817.340bytes)
Webuiltourowntoolstocapturethem,butthereisnopostprocessingonourside,sotheycomerawastheyareread
fromeachEEGdevice,intotal383,751,556DataPoints.
Feelfreetotestanymachinelearning,deeplearningorwhateveralgorithmyouthinkitcouldfit,weonlyaskfor
acknowledgingthesourceandpleaseletusknowofyourperformance!
Wechoosenottodifferentiatethesignalsintotraining/test/validationsetsatthispointsopickthedistributionyouprefer.
Asmallportionofthesignalswerecapturedwithoutthestimulusofseeingthedigitsforcontrast,allarerandomactionsnot
relatedtothinkingorseeingdigits,youcandecidetousethemornotinyourtests,theyusethecode1.
SIGNALDISTRIBUTION:
Thisisthedistributionofthesignalsperdeviceanddigit:
Device/Digit 0 1 2 3 4 5 6 7 8 9 1 Total
MindWave(MW) 5,531 5,498 5,517 5,416 5,381 5,568 5,476 5,552 5,545 5,450 12,701 67,635
EPOC(EP) 91,224 88,914 90,930 92,652 88,886 91,994 91,322 88,718 91,728 91,882 2,226 910,476
Muse(MU) 11,904 11,632 11,920 11,832 11,536 12,052 12,368 12,080 12,208 11,988 44,412 163,932
Insight(IN)* 1.820 1.860 1.805 1.825 1.765 1.905 1.845 1.805 1.690 1.830 0 18.150
Total 110.479 107.904 110.172 111.725 107.568 111.519 111.011 108.155 111.171 111.150 59,3391.160.193
*InsightcapturesstartedinSeptember2015,sosoonwillbeupdatedwithmorebrainsignals,lastupdate09/16/2015
v1.01
FILEFORMAT:
Thedataisstoredinaverysimpletextformatincluding:
[id]:anumeric,onlyforreferencepurposes.
[event]id,ainteger,usedtodistinguishthesameeventcapturedatdifferentbrainlocations,usedonlybymultichannel
devices(allexceptMW).
[device]:a2characterstring,toidentifythedeviceusedtocapturethesignals,"MW"forMindWave,"EP"forEmotive
Epoc,"MU"forInteraxonMuse&"IN"forEmotivInsight.
[channel]:astring,toindentifythe10/20brainlocationofthesignal,withpossiblevalues:
MindWave "FP1"
EPOC "AF3,"F7","F3","FC5","T7","P7","O1","O2","P8","T8","FC6","F4","F8","AF4"
Muse "TP9,"FP1","FP2","TP10"
Insight "AF3,"AF4","T7","T8","PZ"
[code]:ainteger,toindentifythedigitbeenthought/seen,withpossiblevalues0,1,2,3,4,5,6,7,8,9or1forrandom
capturedsignalsnotrelatedtoanyofthedigits.
[size]:ainteger,toidentifythesizeinnumberofvaluescapturedinthe2secondsofthissignal,sincetheHzofeach
16/9/2015 MindBigDatatheMNISTofBrainDigits
http://www.mindbigdata.com/opendb/index.html 2/2
devicevaries,in"theory"thevalueiscloseto512HzforMW,128HzforEP,220HzforMU&128HzforIN,foreachofthe2
seconds.
[data]:acomaseparatedsetofnumbers,withthetimeseriesamplitudeofthesignal,eachdeviceusesadifferent
precisiontoidentifytheelectricalpotentialcapturedfromthebrain:integersinthecaseofMW&MUorrealnumbersinthe
caseofEP&IN.
Thereisnoheadersinthefiles,everylineisasignal,andthefieldsareseparatedbyatab
Forexampleonelineofeachdevicecouldbe(withouttheheaders)
[id] [event] [device] [channel] [code] [size] [data]
27 27 MW FP1 5 952 18,12,13,12,5,3,11,23,37,36,26,24,35,42……
67650 67636 EP F7 7 260 4482.564102,4477.435897,4484.102564…….
978210 132693 MU TP10 1 476 506,508,509,501,497,494,497,490,490,493……
1142043 173652 IN AF3 0 256 4259.487179,4237.948717,4247.179487,4242.051282……
BRAINLOCATIONS:
EachEEGdevicecapturethesignalsviadifferentsensors,locatedintheseareasofmybrain,thecolorrepresentsthe
device:MindWave,EPOC,Muse,Insight
Contactusifyouneedanymoreinfo.
Let'sdecodeMyBrain!
September2015
DavidVivancos
vivancos@vivancos.com
Toward reliable signals decoding for electroencephalogram: A benchmark study to EEGNeX
Article
  • Jan 2024
  • BIOMED SIGNAL PROCES
This study examines the efficacy of various neural network (NN) models in interpreting mental constructs via electroencephalogram (EEG) signals. Through the assessment of 16 prevalent NN models and their variants across four brain-computer interface (BCI) paradigms, we gauged their information representation capability. Rooted in comprehensive literature review findings, we proposed EEGNeX, a novel, purely ConvNet-based architecture. We pitted it against both existing cutting-edge strategies and the Mother of All BCI Benchmarks (MOABB) involving 11 distinct EEG motor imagination (MI) classification tasks and revealed that EEGNeX surpasses other state-of-the-art methods. Notably, it shows up to 2.1%-8.5% improvement in the classification accuracy in different scenarios with statistical significance (p < 0.05) compared to its competitors. This study not only provides deeper insights into designing efficient NN models for EEG data but also lays groundwork for future explorations into the relationship between bioelectric brain signals and NN architectures. For the benefit of broader scientific collaboration, we have made all benchmark models, including EEGNeX, publicly available at (https://github.com/chenxiachan/EEGNeX).
View
View
ResearchGate has not been able to resolve any references for this publication.