An explainable multimodal machine learning model for diagnosing disorders of consciousness: evidence from a large multicenter Chinese cohort.
The 3 matches
- [1] § Methods › EEG features analysis › Spectral analysis ↔ Preprocessing_autoEEG_pipeline.ipynb, lines 69 to 153 · score 0.92 · Power spectral density, 8–13 Hz, parieto occipital, 1–4 Hz, 4–8 Hz, Welch · Report a mismatch
- [2] § Methods › EEG data collection and preprocessing ↔ Preprocessing_autoEEG_pipeline.ipynb, lines 69 to 153 · score 0.85 · eye blink, muscle artifact, ICA components, threshold, Autoreject, epochs · Report a mismatch
- [3] § Methods › EEG data collection and preprocessing ↔ Preprocessing_autoEEG_pipeline.ipynb, lines 22 to 67 · score 0.65 · notch filter, MNE, Preprocessing, resampled, 0.1 Hz, 40 Hz · 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
Jupyter notebook · 161 lines · 5.7 KB · no license · 3 matches
- # %%
- import os
- import mne
- import numpy as np
- import pandas as pd
- from scipy.integrate import simpson
- from autoreject import AutoReject
- from mne.preprocessing import ICA
- from mne_icalabel import label_components
- from mne_connectivity import spectral_connectivity_epochs
- # Folder paths
- folder_path = r'F:\eeg_SHAP' # Enter your folder path
- save_path = r'F:\eeg_SHAP\process' # Enter your save path
- # Get all .vhdr files in the folder
- vhdr_files = [f for f in os.listdir(folder_path) if f.endswith('.vhdr')]
- # Create an empty DataFrame to store features from all files
- all_features = []
- # Iterate over all .vhdr files
- for vhdr_file in vhdr_files:
- file_path = os.path.join(folder_path, vhdr_file)
- file_name = os.path.splitext(vhdr_file)[0]
- # Load EEG data
- raw = mne.io.read_raw_brainvision(file_path, preload=True)
- # Apply standard 1005 electrode layout
- montage = mne.channels.make_standard_montage('standard_1005')
- raw.set_montage(montage)
- # Select only EEG channels
- raw.pick_types(eeg=True)
- # Crop the first 300 seconds of data for processing
- raw.crop(tmin=0, tmax=300)
- raw.filter(l_freq=0.1, h_freq=40.0, picks="eeg")
- raw.notch_filter(freqs=50, notch_widths=4, picks="eeg")
- raw.resample(sfreq=500)
- # Create pseudo-events by segmenting data into 2-second epochs
- interval = 2.0
- n_markers = int(raw.times[-1] // interval)
- events = mne.make_fixed_length_events(raw, id=200, start=0, stop=n_markers * interval, duration=interval)
- # Create epochs object
- epochs = mne.Epochs(
- raw, events, event_id=200, tmin=0, tmax=interval,
- baseline=None, preload=True, detrend=1
- )
- # Use autoreject to automatically handle bad segments and interpolate bad channels
- ar = AutoReject(n_interpolate=[1, 2, 3, 4, 5, 6], random_state=666, n_jobs=-1, verbose=True)
- ar.fit(epochs)
- epochs_clean, reject_log = ar.transform(epochs, return_log=True)
- # ICA preprocessing: filter for training (before fitting ICA)
- raw_for_ica = raw.copy()
- raw_for_ica.filter(l_freq=1.0, h_freq=40.0)
- events_for_ica = mne.make_fixed_length_events(raw_for_ica, id=200, start=0, stop=n_markers * interval, duration=interval)
- epochs_for_ica = mne.Epochs(
- raw_for_ica, events_for_ica, event_id=200, tmin=0, tmax=interval,
- baseline=None, preload=True, detrend=1
- )
- # Use autoreject cleaned epochs indices
- epochs_selection = epochs_clean.selection
- filtered_epochs_for_ica = epochs_for_ica[epochs_selection]
- # ICA fitting
- n_components = min(30, len(filtered_epochs_for_ica.ch_names) - 1)
- ica = ICA(n_components=n_components, max_iter="auto", method="picard", random_state=42, fit_params=dict(ortho=False, extended=True))
- ica.fit(filtered_epochs_for_ica)
- # ICA component labeling
- ic_labels = label_components(filtered_epochs_for_ica, ica, method="iclabel")
- labels = ic_labels["labels"]
- y_pred_proba = ic_labels["y_pred_proba"]
- # Set artifact types and thresholds
- artifact_types = {
- "eye blink": 0.8,
- "heart beat": 0.8,
- "muscle artifact": 0.8
- }
- exclude_idx = []
- for idx, (label, proba_array) in enumerate(zip(labels, y_pred_proba)):
- if label in artifact_types and artifact_types[label] <= np.max(proba_array):
- exclude_idx.append(idx)
- # Apply ICA to remove artifact components
- ica.apply(epochs_clean, exclude=exclude_idx)
- # Remove extreme value epochs
- reject_criteria = dict(eeg=180e-6)
- epochs_clean.drop_bad(reject=reject_criteria)
- # Apply average reference
- epochs_clean.set_eeg_reference(ref_channels='average')
- # Randomly select 30 epochs
- np.random.seed(42)
- max_epochs = 30
- selected_indices = np.random.choice(len(epochs_clean), size=max_epochs, replace=False)
- epochs_ica = epochs_clean[selected_indices]
- # Power spectral analysis - Parieto-occipital alpha relative, delta absolute, alpha/theta ratio
- region = ['Oz', 'O1', 'O2', 'POz', 'PO3', 'PO4', 'PO7', 'PO8', 'P7', 'P5', 'P3', 'P1', 'Pz', 'P2', 'P4', 'P8']
- available = [ch for ch in region if ch in epochs_ica.info['ch_names']]
- psd = epochs_ica.compute_psd(method='welch', fmin=1, fmax=40, picks=available)
- freqs = psd.freqs
- psds = psd.get_data().mean(axis=0).mean(axis=0)
- freq_res = np.mean(np.diff(freqs))
- total = simpson(psds, dx=freq_res)
- def band_power(low, high):
- idx = np.logical_and(freqs >= low, freqs <= high)
- return simpson(psds[idx], dx=freq_res)
- alpha = band_power(8, 13)
- delta = band_power(1, 4)
- theta = band_power(4, 8)
- rel_alpha = alpha / total
- alpha_theta_ratio = alpha / theta
- # Calculate IMCOH theta
- con_theta = spectral_connectivity_epochs(
- epochs_ica, method='imcoh', mode='multitaper', sfreq=epochs_ica.info['sfreq'],
- fmin=4, fmax=8, faverage=True, tmin=0.0, mt_adaptive=False, n_jobs=-1
- ).get_data()
- imcoh_theta = np.mean(np.abs(con_theta))
- # Calculate PLV beta
- con_beta = spectral_connectivity_epochs(
- epochs_ica, method='plv', mode='multitaper', sfreq=epochs_ica.info['sfreq'],
- fmin=13, fmax=30, faverage=True, tmin=0.0, mt_adaptive=False, n_jobs=-1
- ).get_data()
- plv_beta = np.mean(np.abs(con_beta))
- # Save the features for each file
- features = {
- 'File': file_name,
- 'PO_alpha_rel': rel_alpha,
- 'PO_delta_abs': delta * 1e12,
- 'PO_alpha_theta_ratio': alpha_theta_ratio,
- 'PLV_beta': plv_beta,
- 'IMCOH_theta': imcoh_theta
- }
- all_features.append(features)
- # Save features from all files as a CSV
- df_all = pd.DataFrame(all_features)
- output_csv = os.path.join(save_path, 'EEG_features.csv')
- df_all.to_csv(output_csv, index=False)
- print(f"All features saved to: {output_csv}")
Preprocessing_autoEEG_pipeline.ipynb at commit 5d47c46, no license · at the source
Overview
- Department of Neurology, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang China
- School of Medical Technology, Beijing Institute of Technology, Beijing, China
- Centre for Rehabilitation Medicine, Rehabilitation & Sports Medicine Research Institute of Zhejiang Province, Department of Rehabilitation Medicine, Zhejiang Provincial People’s Hospital (Affiliated People’s Hospital, Hangzhou Medical College, Hangzhou, Zhejiang China
- Department of Neurology, Shaoxing Shangyu Second People’s Hospital, Shaoxing, Zhejiang China
- Department of Rehabilitation, Hangzhou Mingzhou Brain Rehabilitation Hospital, Hangzhou, Zhejiang China
- Department of Rehabilitation, Hangzhou Hospital of Zhejiang Armed Police Corps, Hangzhou, Zhejiang China
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 3 matches between paragraphs and lines of code.
Vongoleo/docmodel
5d47c46dec3b8bf113a1c267047df5e0cf805aef, 1 October 2025Availability: 1 check, the latest on 29 September 2026: the link answers
- 29 September 2026: the link answers
3 files
- Predict_shap_analysis_fu
nction.R , R, 100 lines - Preprocessing_autoEEG_pi
peline.ipynb , Jupyter, 161 lines, 3 matches - README.md, Text, 12 lines
The paper's code and data availability statement is in the Data section.
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;
- 2 scripts, each with its path and the digest of its content;
- 3 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
No dataset and no data link were found in the paper.
Code and data availability statement
The paper has a code and data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:
- it points to the authors' code: Vongoleo/
docmodel - it says that the data are available on request
Read it in the paper: doi.org/10.1186/s12967-026-08108-y.
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, 29 September 2026: the first record
Recorded: type, language, journal, volume, issue, pages, dates, 19 authors, 5 keywords, 11 MeSH terms, 5 funders, 33 references.
Cite
This paper
Mou, C., Zhao, J., Yan, Z., Zhang, L., Tian, X., Wang, Y., Wang, H., Hu, J., He, Z., Ling, Y., Kang, A., Luo, Q., Gao, J., Ye, X., Hong, L., Li, J., Yan, T., Yu, J., & Luo, B. (2026). An explainable multimodal machine learning model for diagnosing disorders of consciousness: evidence from a large multicenter Chinese cohort. Journal of translational medicine, 24(1), 708. https://
BibTeX
@article{mou2026explaina
author = {Mou, Chenye and Zhao, Jiajia and Yan, Zilong and Zhang, Li and Tian, Xuejiao and Wang, Yingchen and Wang, Hanxiao and Hu, Jun and He, Zhuolin and Ling, Yi and Kang, Anshun and Luo, Qiwen and Gao, Jian and Ye, Xiangming and Hong, Lirong and Li, Jingqi and Yan, Tianyi and Yu, Jie and Luo, Benyan},
title = {{An explainable multimodal machine learning model for diagnosing disorders of consciousness: evidence from a large multicenter Chinese cohort}},
journal = {Journal of translational medicine},
year = {2026},
month = apr,
volume = {24},
number = {1},
pages = {708},
publisher = {BMC},
issn = {1479-5876},
doi = {10.1186/
url = {https://
pmid = {41987185},
pmcid = {PMC13217752}
}
RIS
TY - JOUR
AU - Mou, Chenye
AU - Zhao, Jiajia
AU - Yan, Zilong
AU - Zhang, Li
AU - Tian, Xuejiao
AU - Wang, Yingchen
AU - Wang, Hanxiao
AU - Hu, Jun
AU - He, Zhuolin
AU - Ling, Yi
AU - Kang, Anshun
AU - Luo, Qiwen
AU - Gao, Jian
AU - Ye, Xiangming
AU - Hong, Lirong
AU - Li, Jingqi
AU - Yan, Tianyi
AU - Yu, Jie
AU - Luo, Benyan
TI - An explainable multimodal machine learning model for diagnosing disorders of consciousness: evidence from a large multicenter Chinese cohort
T2 - Journal of translational medicine
J2 - J Transl Med
PY - 2026
DA - 2026/
VL - 24
IS - 1
SP - 708
SN - 1479-5876
PB - BMC
DO - 10.1186/
UR - https://
LA - en
ER -
CSL-JSON
{
"id": "10.1186/
"type": "article-journal",
"title": "An explainable multimodal machine learning model for diagnosing disorders of consciousness: evidence from a large multicenter Chinese cohort",
"container-title": "Journal of translational medicine",
"author": [
{
"family": "Mou",
"given": "Chenye"
},
{
"family": "Zhao",
"given": "Jiajia"
},
{
"family": "Yan",
"given": "Zilong"
},
{
"family": "Zhang",
"given": "Li"
},
{
"family": "Tian",
"given": "Xuejiao"
},
{
"family": "Wang",
"given": "Yingchen"
},
{
"family": "Wang",
"given": "Hanxiao"
},
{
"family": "Hu",
"given": "Jun"
},
{
"family": "He",
"given": "Zhuolin"
},
{
"family": "Ling",
"given": "Yi"
},
{
"family": "Kang",
"given": "Anshun"
},
{
"family": "Luo",
"given": "Qiwen"
},
{
"family": "Gao",
"given": "Jian"
},
{
"family": "Ye",
"given": "Xiangming"
},
{
"family": "Hong",
"given": "Lirong"
},
{
"family": "Li",
"given": "Jingqi"
},
{
"family": "Yan",
"given": "Tianyi"
},
{
"family": "Yu",
"given": "Jie"
},
{
"family": "Luo",
"given": "Benyan"
}
],
"container-title-short":
"volume": "24",
"issue": "1",
"page": "708",
"DOI": "10.1186/
"PMID": "41987185",
"PMCID": "PMC13217752",
"ISSN": "1479-5876",
"publisher": "BMC",
"URL": "https://
"language": "en",
"issued": {
"date-parts": [
[
2026,
4,
15
]
]
}
}
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:10.1093/brain/awaf412 [code]
- Multimodal multicentre investigation of diagnostic and prognostic markers in disorders of consciousness.Journal: Brain : a journal of neurologyIn common: MNE-Python, pandas, SciPy, 1 other tool, EEG, other condition, 6 references
- [2] doi:10.1007/s12028-023-01816-z [code]
- Multimodal Prediction of 3- and 12-Month Outcomes in ICU Patients with Acute Disorders of Consciousness.Journal: Neurocritical careIn common: autoreject, MNE-Python, pandas, 2 other tools, EEG, other condition, 4 references
- [3] doi:10.1002/hbm.26535 [code]
- How do tasks impact the reliability of fMRI functional connectivity?Journal: Human brain mappingIn common: autoreject, MNE-Connectivity, ICLabel, 5 other tools, 1 reference
- [4] doi:10.1016/j.ebiom.2024.105259 [code]
- Machine learning of brain-specific biomarkers from EEG.Journal: EBioMedicineIn common: autoreject, ICLabel, MNE-Python, 3 other tools, EEG, 3 references
- [5] doi:10.1523/eneuro.0196-22.2023 [code]
- Long-Term Effects of Preterm Birth on Children's Brain Structure: An Analysis of the Adolescent Brain Cognitive Development (ABCD) Study.Journal: eNeuroIn common: autoreject, MNE-Connectivity, ICLabel, 5 other tools, other condition
- [6] doi:10.1162/imag.a.1245 [code]
- Towards precision EEG connectomics: Evaluating the benefits of dense sampling.Journal: Imaging neuroscience (Cambridge, Mass.)In common: autoreject, MNE-Connectivity, ICLabel, 5 other tools, EEG
- [7] doi:10.1162/imag.a.117 [code]
- How much is "enough"? Considerations for functional connectivity reliability in pediatric naturalistic fMRI.Journal: Imaging neuroscience (Cambridge, Mass.)In common: autoreject, MNE-Connectivity, ICLabel, 5 other tools
- [8] doi:10.1016/j.dcn.2023.101268 [code]
- Functional MRI responses to naturalistic stimuli are increasingly typical across early childhood.Journal: Developmental cognitive neuroscienceIn common: autoreject, MNE-Connectivity, ICLabel, 5 other tools
- [9] doi:10.1016/j.dcn.2022.101087 [code]
- Benchmarking common preprocessing strategies in early childhood functional connectivity and intersubject correlation fMRI.Journal: Developmental cognitive neuroscienceIn common: autoreject, MNE-Connectivity, ICLabel, 5 other tools
- [10] doi:10.1038/s41598-024-56633-x [code]
- Gamma power and beta envelope correlation are potential neural predictors of deep hypnosis.Journal: Scientific reportsIn common: autoreject, MNE-Connectivity, MNE-Python, 3 other tools, EEG, cognitive, 2 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.
Claim this paper
Correct its record
Say what each link of this record is, remove the ones that are not the paper's, add the ones that are missing. The correction becomes a new version of the record, in its Versions section.
Validate its tracing map
You validate the map as this page shows it: 1 repository of the authors' code, each at its verified commit and with its license, 2 scripts, and 3 matches between paragraphs and code (see the Code and Map sections). It then receives a DOI on Zenodo, with you (your ORCID iD) and OSCR as its creators; the code itself is not deposited.
The map's fingerprint: sha256:1a3a8454d8185dc1…
Add the badge to the code's README
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 -->
[](https://openscicode.org/badge/doi/10.1186/s12967-026-08108-y)
<!-- oscr-badge:end -->reStructuredText
.. oscr-badge:start
.. image:: https://openscicode.org/badge/doi/10.1186/s12967-026-08108-y.svg
:target: https://openscicode.org/badge/doi/10.1186/s12967-026-08108-y
:alt: OSCR
.. oscr-badge:endHTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1186/s12967-026-08108-y"><img src="https://openscicode.org/badge/doi/10.1186/s12967-026-08108-y.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.
