{"success":true,"database":"eegdash","data":{"_id":"69d16e04897a7725c66f4c53","dataset_id":"ds007558","associated_paper_doi":null,"authors":["Mengsha Qi"],"bids_version":"1.8.0","contact_info":["Mengsha Qi"],"contributing_labs":null,"data_processed":false,"dataset_doi":"doi:10.18112/openneuro.ds007558.v1.0.0","datatypes":["eeg"],"demographics":{"subjects_count":67,"ages":[],"age_min":null,"age_max":null,"age_mean":null,"species":null,"sex_distribution":null,"handedness_distribution":null},"experimental_modalities":null,"external_links":{"source_url":"https://openneuro.org/datasets/ds007558","osf_url":null,"github_url":null,"paper_url":null},"funding":[],"ingestion_fingerprint":"de8078d37cc1fbad3133d2809fe2fccd9e80f106642a5f6f9ff117fde2f5e0f1","license":"CC0","n_contributing_labs":null,"name":"EEG Pre/Post Intervention Dataset","readme":"# Dataset Description\n## Overview\nThis dataset contains EEG recordings from a study investigating neural activity changes before and after an intervention. The data are organized following the Brain Imaging Data Structure (BIDS) specification.\nThe dataset includes multiple participant groups and timepoints:\n- Group 1, Group 2, Group 3\n- Pre-intervention (pre) and Post-intervention (post)\n## Participants\nParticipants are labeled using anonymized IDs (e.g., sub-001, sub-002, etc.). Demographic and session-related information are provided in the corresponding TSV files where applicable.\n## Data Structure\nThe dataset follows the BIDS format:\n- `sub-XXX/`\n  - `ses-pre/` or `ses-post/`\n    - `eeg/`\n      - EEG recordings (.edf)\n      - Metadata files (.json)\n      - Events files (.tsv)\nEach subject contains EEG recordings organized by session (pre/post).\n## Experimental Design\nThe study compares neural activity before and after an intervention. Participants are divided into different groups to evaluate potential differences in outcomes.\n## Data Acquisition\nEEG data were recorded using standard acquisition systems. Detailed acquisition parameters are stored in the accompanying JSON sidecar files.\n## Data Processing\nThe dataset has been reorganized into BIDS format. File naming, metadata, and structure have been standardized to ensure compatibility with BIDS-compliant tools.\n## Known Issues\n- Some warnings may appear during BIDS validation but do not affect data usability.\n- All critical validation errors have been resolved.\n## Usage Notes\nThis dataset can be used for:\n- EEG signal analysis\n- Functional connectivity studies\n- Pre/post intervention comparisons\n## License\nPlease refer to the dataset repository for licensing information.\n## Acknowledgements\nWe thank all participants and researchers involved in data collection and processing.","recording_modality":["eeg"],"senior_author":"Mengsha Qi","sessions":["post","pre"],"size_bytes":719738454,"source":"openneuro","storage":{"backend":"s3","base":"s3://openneuro.org/ds007558","raw_key":"dataset_description.json","dep_keys":["CHANGES","README.md","participants.tsv"]},"study_design":null,"study_domain":null,"tasks":["rest"],"timestamps":{"digested_at":"2026-04-22T12:30:32.825168+00:00","dataset_created_at":"2026-03-25T11:00:03.607Z","dataset_modified_at":"2026-03-25T11:14:07.000Z"},"total_files":121,"computed_title":"EEG Pre/Post Intervention Dataset","nchans_counts":[{"val":19,"count":106},{"val":21,"count":13},{"val":20,"count":2}],"sfreq_counts":[{"val":200.0,"count":121}],"stats_computed_at":"2026-04-22T23:16:00.312955+00:00","total_duration_s":93545.705,"tagger_meta":{"config_hash":"3557b68bca409f28","metadata_hash":"42f21c5b23922fb0","model":"openai/gpt-5.2","tagged_at":"2026-04-07T09:32:40.872789+00:00"},"tags":{"pathology":["Unknown"],"modality":["Resting State"],"type":["Clinical/Intervention"],"confidence":{"pathology":0.45,"modality":0.8,"type":0.85},"reasoning":{"few_shot_analysis":"Most similar conventions in the few-shot set:\n- The dementia dataset (“A dataset of EEG recordings from: Alzheimer's disease, Frontotemporal dementia and Healthy subjects”) is labeled Type=“Clinical/Intervention” even though the recordings are resting-state, because the dataset’s primary purpose is a clinical comparison/biomarker use-case.\n- The sleep deprivation resting dataset is labeled Type=“Resting-state” because it is a non-clinical manipulation and the dataset is framed primarily as resting-state recordings.\nThis target dataset is framed around an “intervention” with pre/post sessions and multiple groups, which follows the convention of using Type=“Clinical/Intervention” when the main organizing principle is an intervention/comparison, even if the task itself is rest.","metadata_analysis":"Key metadata facts (quotes):\n- Purpose framing: “investigating neural activity changes before and after an intervention” and “The study compares neural activity before and after an intervention.”\n- Design structure: “multiple participant groups and timepoints: - Group 1, Group 2, Group 3 - Pre-intervention (pre) and Post-intervention (post)”\n- Task: tasks = [“rest”]\n- Use-case: “Pre/post intervention comparisons” and “Functional connectivity studies”\nNotably missing: no explicit clinical diagnosis/condition (no disease name; groups are unnamed).","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology:\n- Metadata says: only “multiple participant groups” and “Pre-intervention (pre) and Post-intervention (post)”; no diagnosis is named.\n- Few-shot pattern suggests: intervention-style datasets are often clinical, but explicit diagnoses are required for a specific pathology label.\n- Alignment: PARTIAL/UNCLEAR (no explicit pathology fact). Decision relies on absence of diagnostic recruitment info → Pathology=Unknown.\n\nModality:\n- Metadata says: tasks include “rest” and sessions are EEG recordings without described stimuli.\n- Few-shot pattern suggests: when task is rest/eyes open/closed, Modality=“Resting State”.\n- Alignment: ALIGNS strongly → Modality=Resting State.\n\nType:\n- Metadata says: repeated emphasis on “intervention” and “pre/post intervention comparisons” with multiple groups.\n- Few-shot pattern suggests: when the dataset is organized around an intervention/clinical comparison (even with resting recordings), use Type=“Clinical/Intervention” (as in the dementia example).\n- Alignment: ALIGNS → Type=Clinical/Intervention (primary research purpose is intervention effect assessment, not resting-state mapping per se).","decision_summary":"Top-2 candidates and head-to-head:\n\nPathology:\n1) Unknown (WINNER)\n- Evidence: “multiple participant groups” but no condition named; only “before and after an intervention”.\n2) Healthy (RUNNER-UP)\n- Evidence: none explicit; could be a non-clinical intervention in healthy participants, but not stated.\nDecision: Unknown because no explicit recruitment diagnosis or “healthy participants” statement is present.\nConfidence drivers: lack of explicit pathology quotes keeps confidence low.\n\nModality:\n1) Resting State (WINNER)\n- Evidence: tasks = [“rest”]; dataset describes EEG recordings with no stimulus description.\n2) Unknown (RUNNER-UP)\n- Evidence: could be a task mislabeled as rest, but nothing suggests an active paradigm.\nDecision: Resting State.\nConfidence drivers: explicit task label “rest” + repeated pre/post EEG recording framing.\n\nType:\n1) Clinical/Intervention (WINNER)\n- Evidence: “before and after an intervention”; “Pre-intervention (pre) and Post-intervention (post)”; “Pre/post intervention comparisons”.\n2) Resting-state (RUNNER-UP)\n- Evidence: task is rest.\nDecision: Clinical/Intervention because the dataset’s core scientific purpose is intervention-related change, not simply characterizing resting-state activity.\nConfidence drivers: multiple explicit ‘intervention’ quotes."}},"author_year":"Qi2026","canonical_name":null}}