OSCR

Low-power differencing feature extracts spiking-band activities for high-performance intracortical brain-computer interfaces.

Code ↔ Paper

4 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

4 blocks of code here have a permanent identifier (oscr:b:…), listed with the matches below: citable for good, with a page of its own. Select a match and press Cite to copy it. What a block identifier is.

The 4 matches · 1 of them tie a paragraph to a whole file without given lines: a weak match, whose lines are not tinted
  1. [1] § Methods › Neural decoding and performance evaluation ↔ classification_decoding.py, lines 47 to 131 · score 0.71 · Cross Entropy Loss, classifier, fold, concatenate, grasp, models · block oscr:b:MZJRZ0HBK8MR7-0 · Report a mismatch
  2. [2] § Methods › Neural decoding and performance evaluation ↔ Evaluation_of_generalization_performance_and_noise_robustness.py, the whole file · a weak match · score 0.69 · noise robustness, white noise, neural signals, CC, SBP, MAND · block oscr:b:G8ABDM224MDW5-K · Report a mismatch
  3. [3] § Methods › FENet feature extraction and decoding evaluation ↔ FENet_feature_extracting.py, lines 88 to 200 · score 0.63 · FENet, convolutional, dropout, decoder, electrodes, linear · block oscr:b:PZXFZASVX7681-K · Report a mismatch
  4. [4] § Methods › Neural feature extraction ↔ FPGA implementation/python_matlab_verify_sim/feature_ex.py, lines 27 to 34 · score 0.51 · 300–1000 Hz, raw signal, downsampled, bandpass, SBP, 300 Hz · block oscr:b:4Z2S9VHHYTRZ9-N · Report a mismatch

Paper

Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 131 lines · 5.4 KB · GPL-3.0 · 1 match

  1. '''
  2. Related Figures:3d
  3. '''
  4. import os
  5. import torch
  6. import scipy.io
  7. import torch.nn as nn
  8. import numpy as np
  9. from neural_network_model import*
  10. def cut(feature_list,mark_list,lists=0):
  11. '''
  12. Based on mask segmentation features (temporal sequence)
  13. '''
  14. if lists==0:
  15. start_mark=0
  16. end_mark=start_mark
  17. res=[]
  18. while True:
  19. start_mark=end_mark
  20. while True:
  21. end_mark=end_mark+1
  22. if end_mark==len(mark_list) or mark_list[start_mark]!=mark_list[end_mark]:
  23. break
  24. res.append(feature_list[:,start_mark:end_mark])
  25. if end_mark==len(mark_list):
  26. break
  27. else:
  28. start_mark=0
  29. end_mark=start_mark
  30. n=len(mark_list)
  31. m=len(feature_list)
  32. res=[[] for i in range(m)]
  33. while True:
  34. start_mark=end_mark
  35. while True:
  36. end_mark=end_mark+1
  37. if end_mark==n or mark_list[start_mark]!=mark_list[end_mark]:
  38. break
  39. for i in range(m):
  40. res[i].append(feature_list[i][:,start_mark:end_mark])
  41. if end_mark==n:
  42. break
  43. return res
  44. if __name__ == "__main__":
  45. hz_list=['H:/grasp/MUA/']
  46. res_list=['MUA']
  47. dataset_list=['101210.mat','140703.mat']
  48. single_start_path='./single_start_net.pth'
  49. single_best_path='./single_best_net.pth'
  50. classify_start_path='./classify_start_net.pth'
  51. classify_best_path='./classify_best_net.pth'
  52. for w in range(len(hz_list)):
  53. net_model=Classify_Net
  54. loss_fn=nn.CrossEntropyLoss
  55. hidden_size=512
  56. output_size=4
  57. optimizer=torch.optim.Adam
  58. optimizer_kw={'lr':0.001}
  59. best_end_interval=10
  60. for q,dataset in enumerate(dataset_list):
  61. a=scipy.io.loadmat(hz_list[w]+dataset)
  62. a['bined_spk']=((a['bined_spk'].T-a['bined_spk'].mean(1))).T
  63. if q==0:
  64. a['bined_spk'][[1,3,28]]=0
  65. target_num=a['fold_num'].shape[1]
  66. trial_num=a['trial_target'].shape[0]
  67. net_args=[a['bined_spk'].shape[0],hidden_size,output_size]
  68. net_kw={}
  69. loss = np.zeros((trial_num,1))
  70. prediction = np.zeros(trial_num)
  71. single_loss_fn=loss_fn().cuda()
  72. single_net=net_model(*net_args,**net_kw).cuda()
  73. single_optimizer=optimizer(single_net.parameters(),**optimizer_kw)
  74. single_trainer=classify_trainer(single_net,single_optimizer,single_loss_fn)
  75. single_trainer.net_save(classify_start_path)
  76. for i_target in range(a['fold_num'].shape[1]):
  77. target_ind = np.where(a['trial_target']-1 == i_target)[0]
  78. bins_remove = np.concatenate([np.where(a['trial_mask']-1 == target_ind[i])[1] for i in range(len(target_ind))],axis=0)
  79. bined_spk_train=np.delete(a['bined_spk'],bins_remove,axis=1)
  80. trial_mask_train=np.delete(a['trial_mask'],bins_remove,axis=1)
  81. bined_spk_train=cut(bined_spk_train,trial_mask_train[0],lists=0)
  82. label_train=np.delete(a['label'].T,target_ind,axis=0)
  83. for i in range(len(bined_spk_train)):
  84. bined_spk_train[i]=torch.tensor(bined_spk_train[i],dtype=torch.float32).cuda()
  85. label_train=torch.tensor(label_train,dtype=torch.int64).cuda()
  86. bined_spk_test=a['bined_spk'][:,bins_remove]
  87. trial_mask_test=a['trial_mask'][:,bins_remove]
  88. bined_spk_test=cut(bined_spk_test,trial_mask_test[0],lists=0)
  89. label_test=a['label'].T[target_ind]
  90. for i in range(len(bined_spk_test)):
  91. bined_spk_test[i]=torch.tensor(bined_spk_test[i],dtype=torch.float32).cuda()
  92. label_test=torch.tensor(label_test,dtype=torch.int64).cuda()
  93. single_trainer.net_load(classify_start_path)
  94. loss_min=1e10
  95. iteration_best=0
  96. iteration=0
  97. while True:
  98. loss=[]
  99. single_trainer.train_one_turn(bined_spk_train, label_train)
  100. for i in range(len(bined_spk_test)):
  101. los,right=single_trainer.test(bined_spk_test[i], label_test[i])
  102. loss.append(los)
  103. loss_mean=np.array(loss).mean()
  104. if loss_mean<loss_min:
  105. loss_min=loss_mean
  106. iteration_best=iteration
  107. single_trainer.net_save(classify_best_path)
  108. print('{0}-{1}-Loss:{2}'.format(i_target,iteration,loss_mean))
  109. if iteration-iteration_best>best_end_interval:
  110. break
  111. iteration=iteration+1
  112. single_trainer.net_load(classify_best_path)
  113. for i in range(len(target_ind)):
  114. loss[target_ind[i],0],prediction[target_ind[i]] = single_trainer.test(bined_spk_test[i], label_test[i],return_res=1)
  115. create_path=dataset.split('/')[-1].split('.')[0]
  116. os.makedirs('./{0}/{1}'.format(res_list[w],create_path))
  117. np.save('./{0}/{1}/loss.npy'.format(res_list[w],create_path),loss)
  118. np.save('./{0}/{1}/prediction.npy'.format(res_list[w],create_path),prediction)

classification_decoding.py at commit 601fe01, under GPL-3.0 · at the source

Overview

Authors: Guangxiang Xu1,2,3, Chenbin Yu2, Gengrong Shao2, Gang Pan1,4, Yueming Wang2,4, Yaoyao Hao1,2,4
  1. The State Key Lab of Brain-Machine Intelligence, Zhejiang University,Hangzhou, China
  2. Nanhu Brain-computer Interface Institute, Hangzhou, China
  3. Department of Biomedical Engineering, Zhejiang University,Hangzhou, China
  4. College of Computer Science and Technology, Zhejiang University,Hangzhou, China
Institutions: Zhejiang University (China)
Journal: Communications biology, volume 9, issue 1, article 903
Dates: received 1 September 2025; accepted 17 April 2026; published online 29 April 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s42003-026-10144-9 · PMID 42049850 · PMCID PMC13333888 · OpenAlex W7157935271
Open access: gold, a free copy (OpenAlex)
Status: code verified (how the code is found and verified)
Categories: human (organism), methods / tools (subfield)
Methods: Spectral & time-frequency, Connectivity, Statistics, Machine learning, Single-unit activity, calcium imaging, Smoothing, state filtering, decompositions
Keywords: Brain-machine interface, Neural decoding
MeSH: Action Potentials*, Brain-Computer Interfaces*, Signal Processing, Computer-Assisted*, Algorithms, Animals, Handwriting, Humans (* major topic)
Topic: EEG and Brain-Computer Interfaces (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: National Natural Science Foundation of China (62336007, U25D9015, STI2030, 2024C03001); Zhejiang University (SN-ZJU-SIAS-002); Fundamental Research Funds for the Central Universities
Citations: not cited yet (Europe PMC); 46 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repository

Its files are read in the Code ↔ Paper reader above, with 4 matches between paragraphs and lines of code.

Yaoyao-Hao/MAND

License: GPL-3.0
State: the link answers, verified on 30 September 2026
Evidence: files inventoried
Commit: 601fe015c94d17082376068c21ced5960c64c8e5, 7 April 2026
Languages: Stata (28), Shell (24), Python (17), JavaScript (8), MATLAB (4), C (4)
Size: 342 files, 85 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, license file, documentation
Not found: CITATION.cff, environment file, tests, continuous integration
Tools: NumPy (16 files), SciPy (11 files), PyTorch (4 files), Matplotlib (3 files), scikit-learn (2 files)
Availability: 1 check, the latest on 30 September 2026: the link answers
  • 30 September 2026: the link answers
87 files

Code availability statement

The paper has a code availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

Read it in the paper: doi.org/10.1038/s42003-026-10144-9.

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;
  • 85 scripts, each with its path and the digest of its content;
  • 4 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • 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

Datasets cited

Data availability statement

The paper has a data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

  • no repository, dataset or request procedure was recognized in it

Read it in the paper: doi.org/10.1038/s42003-026-10144-9.

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, 30 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 6 authors, 2 keywords, 7 MeSH terms, 3 funders, 40 references.

Cite

This paper

Xu, G., Yu, C., Shao, G., Pan, G., Wang, Y., & Hao, Y. (2026). Low-power differencing feature extracts spiking-band activities for high-performance intracortical brain-computer interfaces. Communications biology, 9(1), 903. https://doi.org/10.1038/s42003-026-10144-9

BibTeX

@article{xu2026low,
author = {Xu, Guangxiang and Yu, Chenbin and Shao, Gengrong and Pan, Gang and Wang, Yueming and Hao, Yaoyao},
title = {{Low-power differencing feature extracts spiking-band activities for high-performance intracortical brain-computer interfaces}},
journal = {Communications biology},
year = {2026},
month = apr,
volume = {9},
number = {1},
pages = {903},
publisher = {Nature Portfolio},
issn = {2399-3642},
doi = {10.1038/s42003-026-10144-9},
url = {https://doi.org/10.1038/s42003-026-10144-9},
pmid = {42049850},
pmcid = {PMC13333888}
}

RIS

TY - JOUR
AU - Xu, Guangxiang
AU - Yu, Chenbin
AU - Shao, Gengrong
AU - Pan, Gang
AU - Wang, Yueming
AU - Hao, Yaoyao
TI - Low-power differencing feature extracts spiking-band activities for high-performance intracortical brain-computer interfaces
T2 - Communications biology
J2 - Commun Biol
PY - 2026
DA - 2026/04/29
VL - 9
IS - 1
SP - 903
SN - 2399-3642
PB - Nature Portfolio
DO - 10.1038/s42003-026-10144-9
UR - https://doi.org/10.1038/s42003-026-10144-9
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s42003-026-10144-9",
"type": "article-journal",
"title": "Low-power differencing feature extracts spiking-band activities for high-performance intracortical brain-computer interfaces",
"container-title": "Communications biology",
"author": [
{
"family": "Xu",
"given": "Guangxiang"
},
{
"family": "Yu",
"given": "Chenbin"
},
{
"family": "Shao",
"given": "Gengrong"
},
{
"family": "Pan",
"given": "Gang"
},
{
"family": "Wang",
"given": "Yueming"
},
{
"family": "Hao",
"given": "Yaoyao"
}
],
"container-title-short": "Commun Biol",
"volume": "9",
"issue": "1",
"page": "903",
"DOI": "10.1038/s42003-026-10144-9",
"PMID": "42049850",
"PMCID": "PMC13333888",
"ISSN": "2399-3642",
"publisher": "Nature Portfolio",
"URL": "https://doi.org/10.1038/s42003-026-10144-9",
"language": "en",
"issued": {
"date-parts": [
[
2026,
4,
29
]
]
}
}

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/s41467-026-70536-7 [code]
Cortical representation of multidimensional handwriting movement and implications for neuroprostheses.
Journal: Nature communications
In common: PyTorch, scikit-learn, SciPy, 1 other tool, 6 references, 3 authors
[2] doi:10.1002/advs.202505492 [code]
Decoding Handwriting Trajectories from Intracortical Brain Signals for Brain-to-Text Communication.
Journal: Advanced science (Weinheim, Baden-Wurttemberg, Germany)
In common: PyTorch, scikit-learn, SciPy, 2 other tools, 12 references, author Yaoyao Hao
[3] doi:10.1088/1741-2552/ad88a4 [code]
Reducing power requirements for high-accuracy decoding in iBCIs.
Journal: Journal of neural engineering
In common: PyTorch, scikit-learn, SciPy, 2 other tools, 10 references
[4] doi:10.1038/s41598-024-51617-3 [code]
Brain control of bimanual movement enabled by recurrent neural networks.
Journal: Scientific reports
In common: scikit-learn, SciPy, NumPy, 9 references
[5] doi:10.1038/s41467-026-71267-5 [code]
Human-like cognitive generalization for large models via mental representation-guided supervision.
Journal: Nature communications
In common: PyTorch, scikit-learn, SciPy, 2 other tools, 2 authors
[6] doi:10.1038/s41591-024-03341-8 [code]
A high-performance brain-computer interface for finger decoding and quadcopter game control in an individual with paralysis.
Journal: Nature medicine
In common: PyTorch, scikit-learn, SciPy, 2 other tools, 8 references
[7] doi:10.7554/elife.89421 [code]
An emerging view of neural geometry in motor cortex supports high-performance decoding.
Journal: eLife
In common: scikit-learn, SciPy, Matplotlib, 1 other tool, 7 references
[8] doi:10.1088/1741-2552/ad038e [code]
From monkeys to humans: observation-basedEMGbrain-computer interface decoders for humans with paralysis.
Journal: Journal of neural engineering
In common: PyTorch, scikit-learn, NumPy, 7 references
[9] doi:10.1038/s41591-026-04414-6 [code]
Long-term independent use of an intracortical brain-computer interface for speech and cursor control.
Journal: Nature medicine
In common: SciPy, Matplotlib, NumPy, 7 references
[10] doi:10.1038/s42003-024-06784-4 [code]
Measuring instability in chronic human intracortical neural recordings towards stable, long-term brain-computer interfaces.
Journal: Communications biology
In common: 8 references

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.

Add the badge to the code's README

OSCR: the badge of this paper

If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.

Markdown
<!-- oscr-badge:start -->
[![OSCR](https://openscicode.org/badge/doi/10.1038/s42003-026-10144-9.svg)](https://openscicode.org/badge/doi/10.1038/s42003-026-10144-9)
<!-- oscr-badge:end -->
reStructuredText
.. oscr-badge:start

.. image:: https://openscicode.org/badge/doi/10.1038/s42003-026-10144-9.svg
   :target: https://openscicode.org/badge/doi/10.1038/s42003-026-10144-9
   :alt: OSCR

.. oscr-badge:end
HTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1038/s42003-026-10144-9"><img src="https://openscicode.org/badge/doi/10.1038/s42003-026-10144-9.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->

On GitHub, open the README, choose Edit (the pencil), paste the snippet under the title, then "Commit changes".

If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge 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: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.

Report a mismatch between the code and the paper · Report a code error (signed in)

Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.

Report a reproduction failure (signed in)

Versions of its code: the releases of its code tied to a version of this paper (the preprint, the accepted manuscript, the version of record…), each with the tracing map versioned with it.

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