OSCR

Deep Learning Discriminates Seizures from Normal Brain Oscillations in the Electroencephalogram of a Rat Model of Post-traumatic Epilepsy.

Code ↔ Paper

The paper beside its authors' code: matches between them have not been computed for this paper yet.

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

MATLAB · 80 lines · 3.8 KB · no license

  2. % Use updated filenames files to locate EEG records, perform multicategory
  3. % detection, and write structured array with detections for each 30 minute file
  4. % updated to handle missing files labeled 'null' as timeline placeholders
  5. clc
  6. clear all;
  8. % Cat: 1=noise, 2=SWDs (r), 3=ESs (g), 4=Spikelets
  9. % Cat: 1 = noise, 2=SWDs (r), 3=spikes (g))
  11. % Experiment = 'Other';
  12. Experiment = 'Mvac';
  13. % Experiment = 'Sean';
  14. % Experiment = 'Jeremy';
  16. [BasePath, RatNames, Chans] = GetRatNames(Experiment);
  17. Nrats = size(RatNames,2);
  19. Ncat=4;
  20. % Ncat=5;
  22. % Fs=500; CutLo=0.1; CutHi=100; Order=1; Notch=1;
  23. Fs=500; CutLo=1; CutHi=40; Order=1; Notch=1; %these are exactly the same as used during training
  24. fb = cwtfilterbank('SignalLength',5000,'VoicesPerOctave',12); %create filterbank in advance for faster repeated application
  25. allImages = zeros(224,224,3,180); %make this once to save a bit of time
  26. allImages = uint8(allImages);
  28. DeletOldDetectionFiles = 1; %use this rarely in case we want to change the training and redo
  29. if(DeletOldDetectionFiles>0)
  30. disp(['Are you sure you want DeletOldDetectionFiles set?']);
  31. pause
  32. end
  33. % for irat = 1:Nrats
  34. for irat =28:28
  35. RatName = char(RatNames(:,irat))
  36. DetectorPath=[BasePath, RatName, '\'];
  37. Channel = Chans(1,irat);
  38. % Channel = 3
  40. load([DetectorPath 'AllFileNames_EEG'],'FileNames');
  41. load([DetectorPath 'GNmodCat',num2str(Ncat)],'trainedGN');
  42. nEEGFiles = size(FileNames,2);
  43. if(nEEGFiles>0)
  44. for ifile=1:nEEGFiles %here is where we process each file for seizures
  45. EEGFileName = deblank(char(FileNames(1,ifile)));
  46. if(strcmp(EEGFileName,'null'));continue;end %just place
  47. % holder for missing file at this date/time so skip
  48. [isfile,isfolder,fbytes,fdate,fdatenum,fpath,fname,fext] = GetFileInfo(EEGFileName);
  49. DETFileFolder = [fpath(1:strfind(fpath,'\EEG')) 'DET\' RatName '\'];
  50. % classifications for each 30 min block are put in DET folder is same folder as EEG as a .mat file with EEG filename
  51. if ~exist(DETFileFolder,'dir');mkdir(DETFileFolder);end %if DET folder does not exist, create it
  52. cd (DETFileFolder);
  53. DETFileName=[fname '.mat'];
  54. if exist (DETFileName, 'file')
  55. if(DeletOldDetectionFiles)
  56. disp('Deleting existent file.')
  57. DETFileName
  58. delete(DETFileName)
  59. end
  60. end
  61. if exist(DETFileName, 'file')
  62. continue
  63. else
  64. % if ~exist(DETFileFolder,'dir');mkdir(DETFileFolder);end %if DET folder does not exist, create it
  65. disp(['Writing DET for ',DETFileFolder, DETFileName])
  66. tic
  67. [LongTrace,ReformData,ReformDataUnfiltered,nlines] = ReadFilterRasterizeLongTraceRaw(EEGFileName,Channel,Fs,CutLo,CutHi,Order,Notch);
  68. % [DETs] = ClassifyMultiCatArray(ReformDataUnfiltered,fb,trainedGN,allImages);
  69. [DETs] = ClassifyMultiCatArray(ReformData,fb,trainedGN,allImages); % We now analize ReformData, which is filtered exactly the same as training EEG
  70. elapsed=toc
  71. nDETs = size(DETs,2);
  72. save (DETFileName, 'DETs')
  73. percentdone=ifile/nEEGFiles*100;
  74. disp(['elepsed time = ',num2str(elapsed),' nDETs = ',num2str(nDETs),' Percent done = ',num2str(percentdone)])
  75. end
  76. end
  77. end
  78. end
  79. % end

Classify.m at commit 53acad6, no license · at the source

Overview

Authors: Sean Tatum1, Jeremy A Taylor1, Katie Waldon1, Anthony J Garcia1, Aaron Witt Jr1, Zachariah Z Smith1, Slavka Ryger1, Andrew Zayachkivsky2, F Edward Dudek2, Daniel S Barth1
ORCID iDs: Jeremy A Taylor, Andrew Zayachkivsky, F Edward Dudek, Daniel S Barth
  1. Department of Psychology and Neuroscience, University of Colorado, Boulder, Colorado 80309
  2. Department of Neurosurgery, University of Utah School of Medicine, Salt Lake City, Utah 84108
Institutions: University of Colorado Boulder (United States); University of Utah (United States)
Journal: eNeuro, volume 13, issue 5, pages ENEURO.0032-26.2026
Dates: received 31 January 2026; accepted 11 April 2026; published online 7 May 2026; in print May 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1523/eneuro.0032-26.2026 · PMID 42045047 · PMCID PMC13160667 · OpenAlex W7156310053
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: EEG (modality), rat (organism), epilepsy (population), systems (subfield)
Methods: Statistics, Preprocessing, Spectral & time-frequency, Physiology & signal measures
Keywords: CNN, SWD
MeSH: Brain*, Deep Learning*, Electroencephalography*, Epilepsy, Post-Traumatic*, Seizures*, Action Potentials, Animals, Brain Waves, Convolutional Neural Networks, Disease Models, Animal, Male, Rats, Rats, Wistar (* major topic)
Topic: Neuroscience and Neuropharmacology Research (Cellular and Molecular Neuroscience, Neuroscience), according to OpenAlex
Citations: not cited yet (Europe PMC); 42 references in the paper

Abstract

This study used machine learning to objectively identify seizures in the electroencephalogram of a model of post-traumatic epilepsy based on fluid percussion injury in male rats. We applied transfer learning to a neural-network trained and tested on three potentially distinct electroencephalographic phenotypes: (1) late-onset convulsive seizures associated with rare post-traumatic epilepsy, (2) early-onset convulsive seizures that often occurred after sham or injury treatment (independent of post-traumatic epilepsy), and (3) spike-wave discharges (SWDs), which occurred in both injured and sham-control rats. The neural network was able to detect seizure events within individual animals and across different cohorts and showed that early and late seizures have similar electroencephalographic phenotypes. Additionally, cross-over training and testing on SWDs from injured and sham-control rats distinguished a convulsive seizure phenotype from normal SWDs. Convolutional neural network modeling of the electroencephalogram can identify spectro-temporal phenotypes that reliably distinguish SWDs from convulsive seizures, indicating that (1) SWDs are normal, not to be falsely classified as nonconvulsive epileptic seizures; (2) the automated detection of convulsive seizures over months revealed rare post-traumatic epilepsy with low seizure frequency; (3) early and late (epileptic) seizures were indistinguishable within and across rats, thus suggesting similar underlying neuronal circuits and ictogenic pathways. This commonality, however, may also obscure important differences between seizure types; (4) convolutional neural network modeling may facilitate objective comparison of seizures within and between laboratories, supplementing subjective expert visual classification, and (5) the rarity of injury-induced epilepsy argues fluid percussion injury is poorly suited for effectively testing anti-epileptogenesis therapies.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above.

dbarth-tech/ML-seizures

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 53acad61302269a1df712e7ac52dab9b4375a482, 19 April 2026
Languages: MATLAB (5)
Size: 8 files, 5 scripts
Software Heritage: not archived
Found in: “Software accessibility”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Signal Processing Toolbox (2 files), Wavelet Toolbox (2 files), Deep Learning Toolbox (1 file), Image Processing Toolbox (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
6 files
At the source: github.com/dbarth-tech/ML-seizures

Software accessibility

All code for CNN modeling and seizure detection was written by the corresponding author, DSB, for use in the common MatLab programming environment. Analysis was performed in MATLAB (R2025a). These program files are available for download at GitHub https://github.com/dbarth-tech/ML-seizures.

Reproduced under the paper's license (CC BY), from the paper cited above.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 5 scripts, each with its path and the digest of its content;
  • no match between paragraphs and code yet;
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data accessibility

These program files are provided along with exemplary EEG data files to evaluate software functionality. Each data file is provided with associated code used for reading it in the correct format.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 10 authors, 2 keywords, 13 MeSH terms, 39 references.

Cite

This paper

Tatum, S., Taylor, J. A., Waldon, K., Garcia, A. J., Witt, A., Smith, Z. Z., Ryger, S., Zayachkivsky, A., Dudek, F. E., & Barth, D. S. (2026). Deep Learning Discriminates Seizures from Normal Brain Oscillations in the Electroencephalogram of a Rat Model of Post-traumatic Epilepsy. eNeuro, 13(5), ENEURO.0032-26.2026. https://doi.org/10.1523/eneuro.0032-26.2026

BibTeX

@article{tatum2026deep,
author = {Tatum, Sean and Taylor, Jeremy A and Waldon, Katie and Garcia, Anthony J and Witt, Aaron and Smith, Zachariah Z and Ryger, Slavka and Zayachkivsky, Andrew and Dudek, F Edward and Barth, Daniel S},
title = {{Deep Learning Discriminates Seizures from Normal Brain Oscillations in the Electroencephalogram of a Rat Model of Post-traumatic Epilepsy}},
journal = {eNeuro},
year = {2026},
month = may,
volume = {13},
number = {5},
pages = {ENEURO.0032--26.2026},
publisher = {Society for Neuroscience},
issn = {2373-2822},
doi = {10.1523/eneuro.0032-26.2026},
url = {https://doi.org/10.1523/eneuro.0032-26.2026},
pmid = {42045047},
pmcid = {PMC13160667}
}

RIS

TY - JOUR
AU - Tatum, Sean
AU - Taylor, Jeremy A
AU - Waldon, Katie
AU - Garcia, Anthony J
AU - Witt, Aaron
AU - Smith, Zachariah Z
AU - Ryger, Slavka
AU - Zayachkivsky, Andrew
AU - Dudek, F Edward
AU - Barth, Daniel S
TI - Deep Learning Discriminates Seizures from Normal Brain Oscillations in the Electroencephalogram of a Rat Model of Post-traumatic Epilepsy
T2 - eNeuro
J2 - eNeuro
PY - 2026
DA - 2026/05/11
VL - 13
IS - 5
SP - ENEURO.0032
EP - 26.2026
SN - 2373-2822
PB - Society for Neuroscience
DO - 10.1523/eneuro.0032-26.2026
UR - https://doi.org/10.1523/eneuro.0032-26.2026
LA - en
ER -

CSL-JSON

{
"id": "10.1523/eneuro.0032-26.2026",
"type": "article-journal",
"title": "Deep Learning Discriminates Seizures from Normal Brain Oscillations in the Electroencephalogram of a Rat Model of Post-traumatic Epilepsy",
"container-title": "eNeuro",
"author": [
{
"family": "Tatum",
"given": "Sean"
},
{
"family": "Taylor",
"given": "Jeremy A"
},
{
"family": "Waldon",
"given": "Katie"
},
{
"family": "Garcia",
"given": "Anthony J"
},
{
"family": "Witt",
"given": "Aaron"
},
{
"family": "Smith",
"given": "Zachariah Z"
},
{
"family": "Ryger",
"given": "Slavka"
},
{
"family": "Zayachkivsky",
"given": "Andrew"
},
{
"family": "Dudek",
"given": "F Edward"
},
{
"family": "Barth",
"given": "Daniel S"
}
],
"container-title-short": "eNeuro",
"volume": "13",
"issue": "5",
"page": "ENEURO.0032-26.2026",
"DOI": "10.1523/eneuro.0032-26.2026",
"PMID": "42045047",
"PMCID": "PMC13160667",
"ISSN": "2373-2822",
"publisher": "Society for Neuroscience",
"URL": "https://doi.org/10.1523/eneuro.0032-26.2026",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
11
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s41467-026-71458-0 [code]
Early differential impact of MeCP2 mutations on functional networks in Rett syndrome patient-derived human cortical organoids.
Journal: Nature communications
In common: Deep Learning Toolbox, Wavelet Toolbox, Image Processing Toolbox, 1 other tool
[2] doi:10.1038/s41593-026-02258-4 [code]
Laminar organization of cellular microcircuits modulating human interictal epileptiform discharges.
Journal: Nature neuroscience
In common: Deep Learning Toolbox, Image Processing Toolbox, Signal Processing Toolbox, epilepsy, systems, EEG
[3] doi:10.3389/fpsyt.2026.1737357 [code]
A computational pipeline for a neurotransmitter-centric analysis of the effects of psychiatric medication on EEG spectral power.
Journal: Frontiers in psychiatry
In common: Deep Learning Toolbox, Wavelet Toolbox, Signal Processing Toolbox, EEG
[4] doi:10.1038/s41598-026-47179-1 [code]
Wavelet-enhanced autoencoder based image denoising with CNN fusion.
Journal: Scientific reports
In common: Deep Learning Toolbox, Wavelet Toolbox, Image Processing Toolbox
[5] doi:10.7554/elife.100642 [code]
Disrupted hippocampal theta-gamma coupling and spike-field coherence following experimental traumatic brain injury.
Journal: eLife
In common: Wavelet Toolbox, Image Processing Toolbox, Signal Processing Toolbox, rat, systems
[6] doi:10.7554/elife.101859 [code]
On-demand seizures facilitate rapid screening of therapeutics for epilepsy.
Journal: eLife
In common: Signal Processing Toolbox, epilepsy, EEG, 2 references
[7] doi:10.1038/s41467-026-76581-6 [code]
Thalamocortical bursts encode reward contingencies and drive associative learning.
Journal: Nature communications
In common: Deep Learning Toolbox, Image Processing Toolbox, Signal Processing Toolbox, systems
[8] doi:10.1371/journal.pbio.3003938 [code]
Theta oscillations tag episodic memories for sleep-dependent consolidation.
Journal: PLoS biology
In common: Wavelet Toolbox, Image Processing Toolbox, Signal Processing Toolbox, EEG
[9] doi:10.1162/imag.a.1229 [code]
40 Hz audiovisual stimulation improves sustained attention and related brain oscillations.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: Wavelet Toolbox, Image Processing Toolbox, Signal Processing Toolbox, EEG
[10] doi:10.1038/s41467-026-77318-1 [code]
Offline generative network reconfiguration guides insight-like accelerated learning by assimilation into schema in rats.
Journal: Nature communications
In common: Wavelet Toolbox, Signal Processing Toolbox, rat, systems

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.