Initialize NeuroAnalyzer
using NeuroAnalyzer
eeg = load("files/eeg.hdf")Basic operations
Editing NEURO objects
Removing object from memory:
eeg = nothingCopying to a new object:
eeg = load("files/eeg.hdf")
eeg_new = deepcopy(eeg)Tip: Do not use eeg_new = eeg as all operations on eeg_new will also affect eeg (see Julia deepcopy() for more details).
All operations and their parameters are stored within the object. Showing the last 5 processing operations:
history(eeg)[1:5]Tip: operations are stored in descending chronological order. Use history(eeg)[end:-1:1] to show the oldest operations first.
Any metadata can be extracted or modified via HEADER structure:
eeg.header.recording[:file_name]"eeg.hdf"Adding a note:
add_note!(eeg,
note = "This is a test description.")Viewing the note:
view_note(eeg)"This is a test description."Deleting note:
delete_note!(eeg)Showing properties:
header(eeg) Data type: EEG
File format: EDF
Source file: eeg.hdf
File size [MB]: 52044.49
Memory size [MB]: 50.8
Subject: 528004 SIT 52, 20220831-122227-{d589f756-53fc-4f1b-915d-6e3b8c1560ad}
Recording: EEGDigiTrack EEG-1042 (42-channel EEG Amplifier) V0.5 Rev. 41
Recording notes:
Recording date: 31.08.22
Recording time: 12:32:53
Sampling rate (Hz): 256
Signal length [samples]: 256001
Signal length [seconds]: 1000.0039
Number of channels: 24
Epochs ID:
Number of epochs: 1
Epoch length [samples]: 256001
Epoch length [seconds]: 1000.0039
Reference type: physical
Line frequency: 50 Hz
Markers: no
Channel locations: yes
Number of EEG channels: 19Showing object properties, all channels details will also be shown (number, label, type, unit and bad status):
info(eeg)Getting basic channels statistics:
NeuroAnalyzer.describe(eeg)< EEG, 24 × 256001 × 1 (1000.00390625 s) >
ch label type unit range mean sd min Q1 median Q3 max
1 Fp1 EEG μV 62.72 0.05 3.4 -32.56 0.0 0.0 5.46 30.16
2 Fp2 EEG μV 50.08 0.08 2.19 -28.64 0.06 0.06 3.54 21.44
3 F3 EEG μV 56.35 0.06 2.87 -32.4 -0.02 -0.02 4.49 23.95
4 F4 EEG μV 80.63 0.05 1.6 -49.64 0.05 0.05 2.37 31.0
5 C3 EEG μV 91.25 0.09 3.85 -50.81 0.02 0.02 6.28 40.44
6 C4 EEG μV 57.68 0.08 2.34 -32.4 0.05 0.05 3.75 25.28
7 P3 EEG μV 70.26 0.06 4.65 -28.04 -0.04 -0.04 7.68 42.22
8 P4 EEG μV 69.95 0.08 4.78 -30.95 0.02 0.02 7.95 38.99
9 O1 EEG μV 70.79 0.1 5.02 -28.39 0.01 0.01 8.35 42.4
10 O2 EEG μV 67.46 0.05 5.12 -28.42 -0.01 -0.01 8.49 39.05
11 F7 EEG μV 100.79 0.07 3.51 -56.62 -0.02 -0.02 5.38 44.17
12 F8 EEG μV 142.67 0.05 2.9 -50.63 0.02 0.02 3.97 92.03
13 T3 EEG μV 67.98 0.09 4.51 -28.14 0.02 0.02 7.47 39.84
14 T4 EEG μV 60.24 0.02 3.93 -29.16 -0.02 -0.02 6.36 31.08
15 T5 EEG μV 71.5 0.09 5.12 -29.82 0.02 0.02 8.51 41.69
16 T6 EEG μV 65.31 0.07 4.69 -28.07 0.03 0.03 7.73 37.25
17 A1 REF μV 111.04 -0.01 6.74 -61.24 -0.05 -0.05 11.0 49.79
18 A2 REF μV 89.31 -0.01 6.44 -52.32 0.08 0.08 10.36 36.99
19 Fz EEG μV 34.46 -0.0 1.65 -23.98 -0.01 -0.01 2.67 10.48
20 Cz EEG μV 60.21 0.12 2.95 -30.13 0.06 0.06 4.82 30.09
21 Pz EEG μV 72.0 0.08 4.51 -34.64 0.02 0.02 7.48 37.36
22 EOG1 EOG μV 196.47 -0.03 7.48 -93.38 -0.05 -0.05 11.38 103.09
23 EOG2 EOG μV 120.71 -0.02 5.38 -64.11 -0.07 -0.07 8.51 56.6
24 ECG ECG mV 1173.84 -0.06 86.37 -615.01 -6.18 -6.18 50.63 558.83 Basic NEURO object properties
Sampling rate:
sr(eeg)256Size of the eeg.data array (channels, samples, epochs):
size(eeg)(24, 256001, 1)Number of channels:
nchannels(eeg)24Number of epochs:
nepochs(eeg)1Signal length in samples:
signal_len(eeg)256001Epoch length in samples:
epoch_len(eeg)256001Channel labels:
l = labels(eeg)
first(l, 5)5-element Vector{String}:
"Fp1"
"Fp2"
"F3"
"F4"
"C3"Object type:
datatype(eeg)"eeg"Converting data to a DataFrame:
df = to_df(eeg)
# show first 5 time points
first(df, 5)5×25 DataFrame
| Row | time | Fp1 | Fp2 | F3 | F4 | C3 | C4 | P3 | P4 | O1 | O2 | F7 | F8 | T3 | T4 | T5 | T6 | A1 | A2 | Fz | Cz | Pz | EOG1 | EOG2 | ECG |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | Float64 | |
| 1 | 0.0 | -3.61876 | 0.451472 | -4.1843 | 3.4805 | 1.29964 | 3.79688 | -0.161454 | 8.19806 | 0.75293 | 5.0857 | -5.62906 | 3.60649 | -2.52252 | 5.05018 | -2.08951 | 6.29989 | -17.5273 | -6.10144 | -1.02441 | 4.27073 | 3.32821 | -6.74063 | -0.963465 | -5.43696 |
| 2 | 0.0039 | -2.62108 | -0.614711 | -1.80555 | 2.54468 | 4.66955 | 4.77973 | 2.88255 | 7.17105 | 3.51212 | 8.42386 | -3.88443 | 3.44589 | 0.997312 | 5.38267 | 2.13859 | 6.55183 | -10.4832 | -0.887447 | -0.871635 | 4.09755 | 5.51992 | -4.54914 | -2.07689 | -8.15171 |
| 3 | 0.0078 | -0.935175 | -3.5474 | 0.853763 | 0.369701 | 5.54463 | 3.4151 | 5.49478 | 8.89833 | 7.43718 | 10.4673 | -0.3261 | 2.15743 | 4.76706 | 4.92332 | 5.87302 | 7.32285 | -3.34778 | 1.38792 | -0.461632 | 4.09585 | 7.44717 | -1.63285 | -1.86917 | -0.795649 |
| 4 | 0.0117 | -3.67928 | -1.21269 | -5.3538 | 1.62046 | -2.76407 | 1.25433 | -3.43964 | 5.69924 | -1.98902 | 3.70845 | -6.4243 | 3.0027 | -3.31217 | 2.61098 | -3.35343 | 2.86376 | -13.1599 | -4.03449 | -0.873815 | -0.640716 | 0.35095 | -2.73419 | 3.86397 | -10.4405 |
| 5 | 0.0156 | -2.24259 | 0.535407 | -2.63971 | 1.72694 | -2.19496 | 0.146516 | -4.57769 | 1.1243 | -3.62526 | 2.80111 | -2.96658 | 0.375008 | -4.23205 | 0.52774 | -4.10575 | 0.936797 | -5.97518 | -1.0515 | -0.895591 | -2.582 | -1.36012 | -6.01542 | 3.53709 | -4.47885 |
or
df = info(eeg, df = true)Channel operations
Showing channel labels:
labels(eeg)24-element Vector{String}:
"Fp1"
"Fp2"
"F3"
"F4"
"C3"
"C4"
"P3"
"P4"
"O1"
"O2"
"F7"
"F8"
"T3"
"T4"
"T5"
"T6"
"A1"
"A2"
"Fz"
"Cz"
"Pz"
"EOG1"
"EOG2"
"ECG"Getting channel by its label:
get_channel(eeg, ch = "Fp1")1-element Vector{Int64}:
1Renaming channel label:
rename_channel(eeg,
ch = "Cz",
name = "CZ")Deleting channels (epochs and channels may be specified using number, range or vector):
delete_channel(eeg, ch = "Cz")Deleting a range of channels:
chs = labels(obj)[10:18]
delete_channel(eeg, ch = chs)Keeping channel(s) - will delete all other channels:
keep_channel(eeg,
ch = get_channel(eeg, type = "eeg")[1:4])Replacing channel 1 data with channel 18 data:
ch1 = 1
ch2 = 18
ch_data = extract_channel(eeg,
ch = get_channel(eeg, type = "all")[ch2])
replace_channel(eeg,
ch = get_channel(eeg, type = "all")[1],
s = ch_data)Epoch operations
Epoching is a procedure in which specific time-windows are extracted from the continuous EEG signal; these time windows are called epochs. Epochs usually are time-locked with respect an event e.g. a visual stimulus. Recorded data is stored as matrix channel × time (where time is the complete continuous the signal). After the epoching procedure the matrix is transformed into an array: channel × time × epochs, where time is the time length of each epoch, and epochs is the number of segments you extracted from continuous signal.
Imagine we have a signal with length of 60 s and our sampling frequency is 1 Hz. The matrix representation of your EEG signal would be 1×60 array or matrix, so if you divide your main signal to some 2 s signals, you would have 30 epoch (each 2 s of your main signal would be an epoch). This could be done by a simple for loop resulting in a new matrix which size is 30×1×2.
If you want to extract epochs from your signal, you should know what are the segments of interest to be analyzed, for instance, a specific stimulus.
Tip: Many processing operations (such as filtering or ICA decomposition) should be performed on non-epoched (continuous) signal (e.g. due to the edge artifacts that may occur in case of filtering or the minimum required signal length for ICA decomposition).
Epoching EEG into 10-second epochs:
e10 = epoch(eeg, ep_len = 10)
header(e10) Data type: EEG
File format: EDF
Source file: eeg.hdf
File size [MB]: 52044.49
Memory size [MB]: 48.87
Subject: 528004 SIT 52, 20220831-122227-{d589f756-53fc-4f1b-915d-6e3b8c1560ad}
Recording: EEGDigiTrack EEG-1042 (42-channel EEG Amplifier) V0.5 Rev. 41
Recording notes:
Recording date: 31.08.22
Recording time: 12:32:53
Sampling rate (Hz): 256
Signal length [samples]: 256000
Signal length [seconds]: 1000.0
Number of channels: 24
Epochs ID: length_10s
Number of epochs: 100
Epoch length [samples]: 2560
Epoch length [seconds]: 10.0
Reference type: physical
Line frequency: 50 Hz
Markers: no
Channel locations: yes
Number of EEG channels: 19Getting data of the 1st epoch:
e10e1 = extract_epoch(e10, ep = 1)
header(e10e1) Data type: EEG
File format: EDF
Source file: eeg.hdf
File size [MB]: 52044.49
Memory size [MB]: 0.51
Subject: 528004 SIT 52, 20220831-122227-{d589f756-53fc-4f1b-915d-6e3b8c1560ad}
Recording: EEGDigiTrack EEG-1042 (42-channel EEG Amplifier) V0.5 Rev. 41
Recording notes:
Recording date: 31.08.22
Recording time: 12:32:53
Sampling rate (Hz): 256
Signal length [samples]: 2560
Signal length [seconds]: 10.0
Number of channels: 24
Epochs ID: length_10s
Number of epochs: 1
Epoch length [samples]: 2560
Epoch length [seconds]: 10.0
Reference type: physical
Line frequency: 50 Hz
Markers: no
Channel locations: yes
Number of EEG channels: 19Deleting epochs:
delete_epoch(e10, ep = 1)
delete_epoch(e10, ep = 8:10)
delete_epoch(e10, ep = [1:5; 9; 10])Tip: Note the use of ; if a noncontinuous range is specified.
Keeping epochs (deleting all other epochs):
keep_epoch(e10,
ep = [1, 3, 5, 9])Split into 5-second epochs and then average into one epoch:
e5 = epoch(eeg, ep_len = 5)
e5avg = average_epochs(e5)
header(e5avg) Data type: ERP
File format: EDF
Source file: eeg.hdf
File size [MB]: 52044.49
Memory size [MB]: 39.29
Subject: 528004 SIT 52, 20220831-122227-{d589f756-53fc-4f1b-915d-6e3b8c1560ad}
Recording: EEGDigiTrack EEG-1042 (42-channel EEG Amplifier) V0.5 Rev. 41
Recording notes:
Recording date: 31.08.22
Recording time: 12:32:53
Sampling rate (Hz): 256
Signal length [samples]: 257280
Signal length [seconds]: 1005.0
Number of channels: 19
Epochs ID: length_5s
Number of epochs: 201
Epoch length [samples]: 1280
Epoch length [seconds]: 5.0
Markers: no
Channel locations: yes
Number of EEG channels: 19Tip: average_epochs() averages EEG/MEG epochs. Non-EEG/MEG channels are removed. OBJ.header.recording[:data_type] becomes erp (for EEG) or erf for MEG. First epoch is the ERP/ERF.
Epoching at the event marker (each epoch is 1500 ms long and starts 200 ms before the event marker):
erp = epoch(eeg,
marker = "Eyes Open",
offset = 0.2,
ep_len = 1.5)Creating sub-epochs (of reduced time range, starting at 0.1 sec and ending at 1.1 sec):
subepoch(erp,
ep_start = 0.1,
ep_end = 1.1)