NM000398: ieeg dataset, 4 subjects#
Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids
Access recordings and metadata through EEGDash.
Citation: Mrinal Pahwa, Matthew Kusner, Carl D. Hacker, David T. Bundy, Kilian Q. Weinberger, Eric C. Leuthardt (2015). Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids. 10.82901/nemar.nm000398
Modality: ieeg Subjects: 4 Recordings: 23 License: CC0-1.0 Source: nemar
Metadata: Complete (100%)
4-participant iEEG dataset — Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids.
Quickstart#
Install
pip install eegdash
Access the data
from eegdash.dataset import NM000398
dataset = NM000398(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)
Filter by subject
dataset = NM000398(cache_dir="./data", subject="01")
Advanced query
dataset = NM000398(
cache_dir="./data",
query={"subject": {"$in": ["01", "02"]}},
)
Iterate recordings
for rec in dataset:
print(rec.subject, rec.raw.info['sfreq'])
If you use this dataset in your research, please cite the original authors.
BibTeX
@dataset{nm000398,
title = {Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids},
author = {Mrinal Pahwa and Matthew Kusner and Carl D. Hacker and David T. Bundy and Kilian Q. Weinberger and Eric C. Leuthardt},
doi = {10.82901/nemar.nm000398},
url = {https://doi.org/10.82901/nemar.nm000398},
}
About This Dataset#
Subdural ECoG from four patients with intractable epilepsy (Barnes Jewish Hospital, St. Louis) with 8x8 grids over left
frontal, temporal and parietal cortex. For each patient the release gives two “wakeful” and two “sleep-like” epochs, identified from video/audio of natural behaviour during clinical monitoring (no task).
wakeful and sleep-like states for future electrocorticographic brain computer interface applications.
doi:10.5061/dryad.4f92n (version 1, 2016-01-26). License: CC0 1.0 (Dryad).
Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015)
Article: PLoS One 10(11):e0142947 (2015), doi:10.1371/journal.pone.0142947 (open access, PMC4643046).
All 24 Dryad files were downloaded through the Dryad API and matched the Dryad md5 digests and sizes.
Sampling rate (important)
The .mat files contain only a 64-column ‘data’ matrix: no sampling rate, channel names or units. The sampling rate
View full README
Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015)
Article: PLoS One 10(11):e0142947 (2015), doi:10.1371/journal.pone.0142947 (open access, PMC4643046).
All 24 Dryad files were downloaded through the Dryad API and matched the Dryad md5 digests and sizes.
Sampling rate (important)
The .mat files contain only a 64-column ‘data’ matrix: no sampling rate, channel names or units. The sampling rate used here comes from the release README (README_for_SubA_Sleep1.pdf): subject A 256 Hz, subjects B, C and D 512 Hz.
The article’s Methods say the signals were “sampled at 256 Hz” for all patients. We follow the README, which is specific to the released files; it is not verifiable from the files themselves.
Contents
sub-<A-D>/ieeg/sub-<X>_task-<wake|sleep>[_acq-part<k>]_run-<n>_ieeg.*: 23 files, 13.62 h in total at the README rates.run= epoch number of the release (Sleep1/Sleep2/Wake1/Wake2). Epochs released as two files (_part1,_part2) are kept as two files (acq-part1,acq-part2); the part lengths differ by at most one sample, consistent with one recording cut in half, but the release does not say so. Mean absolute sample-to-sample step at the junction vs typical step (µV): A_sleep2: 12.1 vs 11.6; A_wake1: 15.0 vs 27.7; A_wake2: 21.7 vs 36.5; B_sleep2: 74.7 vs 78.1; B_wake2: 33.5 vs 77.8; C_sleep2: 41.6 vs 43.8; D_sleep1: 33.6 vs 46.8.Channels
ch01..``ch64`` = columns 1-64 of the release matrix, typed ECOG. Electrode positions are shown only as images in the README PDF (kept in sourcedata); no coordinates or labels are released.Several files have many samples at the amplitude limit |x| = 5482.29 (saturation), mostly subjects B and D (more than half of the samples of at least one channel in: SubB_Sleep1.mat, SubB_Sleep2_part1.mat, SubB_Sleep2_part2.mat, SubB_Wake1.mat, SubB_Wake2_part1.mat, SubB_Wake2_part2.mat, SubD_Sleep2.mat, SubD_Wake2.mat). The percentage per channel is in the
descriptioncolumn of each channels.tsv. Channels are not marked bad: the release has no channel quality information.sourcedata/dryad-4f92n-deidentified/: all released files (the 23 .mat files and README_for_SubA_Sleep1.pdf); the only change is that MAT text-header and PDF metadata dates are reduced to month and year (day -> 01).DEIDENTIFICATION_MANIFEST.tsvlists original and new sha-256.
Conversion
Values are written as BrainVision IEEE_FLOAT_32; every value of the float64 source is exactly representable in float32 (checked per file), so values are identical. MNE read-back matches the source.
The release states the signals are raw, with only the amplifiers’ 0.1 Hz hardware high-pass. The unit is not stated; µV is assumed (amplitudes are consistent with µV).
Participant age, sex, handedness, seizure foci: article Table 1 (subject letters A-D match the release).
Privacy
No names, dates of recording or hospital identifiers in the files. The README PDF names the corresponding author (contact e-mail), not patients. File-creation dates (MAT header, PDF metadata) reduced to month and year.
Additional metadata and localisation (added 2026-10-08)
Compiled after the upload from the article, its supplement and the source deposit (each statement names its source). Text and sidecar metadata only; no data file was changed.
Sources: P = Pahwa, Kusner, Hacker, Bundy, Weinberger, Leuthardt 2015, PLoS ONE 10(11):e0142947, doi:10.1371/journal.pone.0142947 (PMC4643046). R = deposit README_for_SubA_Sleep1.pdf (one page). Reference. All electrodes were referenced to a skull-facing electrode of the same size (P, Methods). The analysis also regressed out the mean of non-noisy electrodes (P). Electrodes. PMT subdural grids, 8x8, flat circular platinum electrodes of 2.3 mm diameter with 10 mm spacing, over the left frontal, temporal and parietal cortex (P, Methods and Fig 2). Localisation. The cortex was reconstructed from the pre-op T1 and electrodes located on the post-op CT. Electrodes and surface were co-registered to a common atlas space and projected onto the pial surface along the grid normal, keeping 10 mm spacing (after Hermes et al. 2010) (P, “Construction of Subject-averaged Cortical Maps”). The electrode positions for each subject appear only as figures (P Fig 2A; R). No coordinates or per-electrode anatomical labels were deposited, and the file channel order is not linked to grid position.
NEMAR Metadata#
[](https://doi.org/10.82901/nemar.nm000398) # Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015) Subdural ECoG from four patients with intractable epilepsy (Barnes Jewish Hospital, St. Louis) with 8x8 grids over left frontal, temporal and parietal cortex. For each patient the release gives two “wakeful” and two “sleep-like” epochs, identified from video/audio of natural behaviour during clinical monitoring (no task). ## Source - Dryad: Pahwa M, Kusner M, Hacker CD, Bundy DT, Weinberger KQ, Leuthardt EC. Data from: Optimizing the detection of
wakeful and sleep-like states for future electrocorticographic brain computer interface applications. doi:10.5061/dryad.4f92n (version 1, 2016-01-26). License: CC0 1.0 (Dryad).
Article: PLoS One 10(11):e0142947 (2015), doi:10.1371/journal.pone.0142947 (open access, PMC4643046).
All 24 Dryad files were downloaded through the Dryad API and matched the Dryad md5 digests and sizes.
## Sampling rate (important) The .mat files contain only a 64-column ‘data’ matrix: no sampling rate, channel names or units. The sampling rate used here comes from the release README (README_for_SubA_Sleep1.pdf): subject A 256 Hz, subjects B, C and D 512 Hz. The article’s Methods say the signals were “sampled at 256 Hz” for all patients. We follow the README, which is specific to the released files; it is not verifiable from the files themselves. ## Contents - sub-<A-D>/ieeg/sub-<X>_task-<wake|sleep>[_acq-part<k>]_run-<n>_ieeg.*: 23 files, 13.62 h in
total at the README rates. run = epoch number of the release (Sleep1/Sleep2/Wake1/Wake2). Epochs released as two files (_part1, _part2) are kept as two files (acq-part1, acq-part2); the part lengths differ by at most one sample, consistent with one recording cut in half, but the release does not say so. Mean absolute sample-to-sample step at the junction vs typical step (µV): A_sleep2: 12.1 vs 11.6; A_wake1: 15.0 vs 27.7; A_wake2: 21.7 vs 36.5; B_sleep2: 74.7 vs 78.1; B_wake2: 33.5 vs 77.8; C_sleep2: 41.6 vs 43.8; D_sleep1: 33.6 vs 46.8.
Channels ch01..`ch64` = columns 1-64 of the release matrix, typed ECOG. Electrode positions are shown only as images in the README PDF (kept in sourcedata); no coordinates or labels are released.
Several files have many samples at the amplitude limit |x| = 5482.29 (saturation), mostly subjects B and D (more than half of the samples of at least one channel in: SubB_Sleep1.mat, SubB_Sleep2_part1.mat, SubB_Sleep2_part2.mat, SubB_Wake1.mat, SubB_Wake2_part1.mat, SubB_Wake2_part2.mat, SubD_Sleep2.mat, SubD_Wake2.mat). The percentage per channel is in the description column of each channels.tsv. Channels are not marked bad: the release has no channel quality information.
sourcedata/dryad-4f92n-deidentified/: all released files (the 23 .mat files and README_for_SubA_Sleep1.pdf); the only change is that MAT text-header and PDF metadata dates are reduced to month and year (day -> 01). DEIDENTIFICATION_MANIFEST.tsv lists original and new sha-256.
## Conversion - Values are written as BrainVision IEEE_FLOAT_32; every value of the float64 source is exactly representable in
float32 (checked per file), so values are identical. MNE read-back matches the source.
The release states the signals are raw, with only the amplifiers’ 0.1 Hz hardware high-pass. The unit is not stated; µV is assumed (amplitudes are consistent with µV).
Participant age, sex, handedness, seizure foci: article Table 1 (subject letters A-D match the release).
## Privacy - No names, dates of recording or hospital identifiers in the files. The README PDF names the corresponding author
(contact e-mail), not patients. File-creation dates (MAT header, PDF metadata) reduced to month and year.
## Additional metadata and localisation (added 2026-10-08) Compiled after the upload from the article, its supplement and the source deposit (each statement names its source). Text and sidecar metadata only; no data file was changed. Sources: P = Pahwa, Kusner, Hacker, Bundy, Weinberger, Leuthardt 2015, PLoS ONE 10(11):e0142947, doi:10.1371/journal.pone.0142947 (PMC4643046). R = deposit README_for_SubA_Sleep1.pdf (one page). Reference. All electrodes were referenced to a skull-facing electrode of the same size (P, Methods). The analysis also regressed out the mean of non-noisy electrodes (P). Electrodes. PMT subdural grids, 8x8, flat circular platinum electrodes of 2.3 mm diameter with 10 mm spacing, over the left frontal, temporal and parietal cortex (P, Methods and Fig 2). Localisation. The cortex was reconstructed from the pre-op T1 and electrodes located on the post-op CT. Electrodes and surface were co-registered to a common atlas space and projected onto the pial surface along the grid normal, keeping 10 mm spacing (after Hermes et al. 2010) (P, “Construction of Subject-averaged Cortical Maps”). The electrode positions for each subject appear only as figures (P Fig 2A; R). No coordinates or per-electrode anatomical labels were deposited, and the file channel order is not linked to grid position.
License: CC0-1.0
Authors:
Mrinal Pahwa
Matthew Kusner
Carl D. Hacker
David T. Bundy
Kilian Q. Weinberger
… and 1 more
Versions:
Version |
DOI |
Released |
|---|---|---|
|
Cohort#
Dataset Statistics#
Age distribution by gender (n=4, range 21–55 yr, mean 37.2 yr)
Sex composition
Channel counts: 64 ch (n=23 recordings)
Sampling frequencies (Hz)
Total recording duration: 13 h 37 min
Signal · Electrodes & live trace#
Live trace viewer — sub-A · task-sleep · run-2
Showing one representative recording out of
4 subjects and 23 recordings in this dataset.
Browse the full set on OpenNeuro;
drop any other _ieeg.{set,edf,bdf,vhdr} file onto the
viewer (or pass ?ieeg=<url>) to inspect it.
No scalp electrode layout is currently indexed for this dataset. Once the eegdash montage registry ingests it, the interactive viewer will appear here automatically.
NEMAR Processing Statistics#
The plots below are generated by NEMAR’s automated EEG pipeline. The histogram shows pipeline success for data cleaning and ICA decomposition, the percentage of data frames and EEG channels retained after artefact removal, line noise per channel (RMS, dB), and the age/gender distribution of participants.
HED event descriptors word cloud
Manifest#
File Explorer#
Browse the BIDS file structure of this dataset. Records are fetched on demand from the EEGDash catalog the first time you open the explorer.
Full dataset metadata table
Dataset ID |
|
Title |
Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids |
Author (year) |
— |
Canonical |
— |
Importable as |
|
Year |
2015 |
Authors |
Mrinal Pahwa, Matthew Kusner, Carl D. Hacker, David T. Bundy, Kilian Q. Weinberger, Eric C. Leuthardt |
License |
CC0-1.0 |
Citation / DOI |
|
Source links |
OpenNeuro | NeMAR | Source URL |
Copy-paste BibTeX
@dataset{nm000398,
title = {Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids},
author = {Mrinal Pahwa and Matthew Kusner and Carl D. Hacker and David T. Bundy and Kilian Q. Weinberger and Eric C. Leuthardt},
doi = {10.82901/nemar.nm000398},
url = {https://doi.org/10.82901/nemar.nm000398},
}
API Reference#
eegdash.datasetEEGDashDataset- class eegdash.dataset.NM000398(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#
Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids
- Study:
nm000398(NeMAR)- Author (year):
—
- Canonical:
—
Also importable as:
NM000398.Modality:
ieeg; Subject type:Unknown. Subjects: 4; recordings: 23; tasks: 2.- Parameters:
cache_dir (str | Path) – Directory where data are cached locally.
query (dict | None) – Additional MongoDB-style filters to AND with the dataset selection. Must not contain the key
dataset.s3_bucket (str | None) – Base S3 bucket used to locate the data.
**kwargs (dict) – Additional keyword arguments forwarded to
EEGDashDataset.
- data_dir#
Local dataset cache directory (
cache_dir / dataset_id).- Type:
Path
Notes
Each item is a recording; recording-level metadata are available via
dataset.description.querysupports MongoDB-style filters on fields inALLOWED_QUERY_FIELDSand is combined with the dataset filter. Dataset-specific caveats are not provided in the summary metadata.References
OpenNeuro dataset: https://openneuro.org/datasets/nm000398 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000398 DOI: https://doi.org/10.82901/nemar.nm000398
Examples
>>> from eegdash.dataset import NM000398 >>> dataset = NM000398(cache_dir="./data") >>> recording = dataset[0] >>> raw = recording.load()
- __init__(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#
- save(path: str, overwrite: bool = False, offset: int = 0)[source]#
Save datasets to files by creating one subdirectory for each dataset:
path/ 0/ 0-raw.fif | 0-epo.fif description.json raw_preproc_kwargs.json (if raws were preprocessed) window_kwargs.json (if this is a windowed dataset) window_preproc_kwargs.json (if windows were preprocessed) target_name.json (if target_name is not None and dataset is raw) 1/ 1-raw.fif | 1-epo.fif description.json raw_preproc_kwargs.json (if raws were preprocessed) window_kwargs.json (if this is a windowed dataset) window_preproc_kwargs.json (if windows were preprocessed) target_name.json (if target_name is not None and dataset is raw)
- Parameters:
path (str) –
- Directory in which subdirectories are created to store
-raw.fif | -epo.fif and .json files to.
overwrite (bool) – Whether to delete old subdirectories that will be saved to in this call.
offset (int) – If provided, the integer is added to the id of the dataset in the concat. This is useful in the setting of very large datasets, where one dataset has to be processed and saved at a time to account for its original position.
BaseDataset from braindecode — windowed via create_windows_from_events.braindecodeDataLoader; supports parallel workers and on-the-fly augmentations.pytorch
Find datasets with the EEGDash APIQuery the catalogue, filter by task or modality, list candidates.
Load one EEG recordingResolve a single record to an MNE Raw with channels and events.
EEG recording to PyTorch DataLoaderWrap braindecode windows in a DataLoader for model training.
Preprocess EEG and create windowsFilter, resample, epoch — and persist the windowed dataset.
Save and reload prepared dataCache a windowed dataset to disk and reattach it without recompute.
Download a dataset locallyPrefetch BIDS files to a local cache and validate the layout.Swap any load_dataset(...) call for nm000398 to reproduce the tutorial on this dataset.
Citation
Mrinal Pahwa, Matthew Kusner, Carl D. Hacker, David T. Bundy, Kilian Q. Weinberger, … (2015). Wakeful and sleep-like states in subdural ECoG (Pahwa et al., 2015): 4 patients, 64-channel grids. 10.82901/nemar.nm000398
Provenance
¹Contributed to nemar in BIDS format.
²Curated & ingested by the EEGDash catalog; see CITATION.cff for canonical reference.
³Persistent identifier: 10.82901/nemar.nm000398.
See Also#
eegdash.dataset.EEGDashDataseteegdash.dataset