{"success":true,"database":"eegdash","data":{"_id":"6953f4239276ef1ee07a32bc","dataset_id":"ds003078","associated_paper_doi":null,"authors":["Philippe DOMENECH","Sylvain RHEIMS","Etienne KOECHLIN"],"bids_version":"1.0.0","contact_info":["philippe Domenech"],"contributing_labs":null,"data_processed":false,"dataset_doi":"10.18112/openneuro.ds003078.v1.0.0","datatypes":["ieeg"],"demographics":{"subjects_count":6,"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/ds003078","osf_url":null,"github_url":null,"paper_url":null},"funding":["European Research Council Grant ERC-AdG #250106","IHU CESAME program “Investissements d’Avenir” ANR-10-IBHU-0003"],"ingestion_fingerprint":"cb48128625413fe4fc5f6a3e6ef27777af7eefd5451eff1cd35556536b39c689","license":"CC0","n_contributing_labs":null,"name":"PROBE iEEG","readme":"--- version 1.0.0 ---\ninitial release\nRaw iEEG data + pre and post surgery MRI\nNo reference to the published study (still in press)","recording_modality":["ieeg"],"senior_author":"Etienne KOECHLIN","sessions":[],"size_bytes":11761361020,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["PROBE"],"timestamps":{"digested_at":"2026-04-22T12:25:31.049476+00:00","dataset_created_at":"2020-08-16T19:12:06.282Z","dataset_modified_at":"2020-08-17T04:11:36.000Z"},"total_files":72,"storage":{"backend":"s3","base":"s3://openneuro.org/ds003078","raw_key":"dataset_description.json","dep_keys":["CHANGES","README"]},"tagger_meta":{"config_hash":"3557b68bca409f28","metadata_hash":"8851237d3b805f3a","model":"openai/gpt-5.2","tagged_at":"2026-04-07T09:32:40.872789+00:00"},"tags":{"pathology":["Surgery"],"modality":["Unknown"],"type":["Unknown"],"confidence":{"pathology":0.7,"modality":0.4,"type":0.4},"reasoning":{"few_shot_analysis":"Most similar few-shot by population/context is the pediatric epilepsy dataset (Pathology=Epilepsy) because it is also clearly clinical and involves specialized EEG acquisition for a medical purpose (sleep EEG with HFO marking). However, unlike that example, this dataset’s metadata does not explicitly say “epilepsy”; it only indicates an invasive recording context tied to surgery. The few-shot conventions suggest: (a) when a clinical population is explicit, label the pathology accordingly; (b) when the dataset focus is clinical/medical, Type often maps to Clinical/Intervention. Because the current metadata lacks explicit diagnosis and task/stimulus details, I follow the convention of not inferring a specific disorder unless stated, and instead use the more general recruited condition available in metadata (“surgery”).","metadata_analysis":"Key available metadata is sparse. Relevant quoted facts:\n1) Readme: \"Raw iEEG data + pre and post surgery MRI\".\n2) Readme: \"No reference to the published study (still in press)\" (no task/paradigm details provided).\n3) Participants: \"Subjects: 6\" (no demographics/diagnosis).\nTask list only contains: \"PROBE\" with no TaskDescription, and there is no mention of sensory stimuli (visual/auditory/etc.) or resting/sleep.","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology:\n- Metadata says: \"pre and post surgery MRI\" and \"Raw iEEG data\" (invasive recordings in a surgical context).\n- Few-shot pattern suggests that invasive/clinical datasets often correspond to a specific disorder when explicitly stated (e.g., epilepsy), but also that we should not guess a diagnosis if it is not written.\n- Alignment: partial. Both indicate a clinical context, but metadata does not name a diagnosis. Therefore, choose Surgery as the most directly supported recruited clinical condition.\n\nModality:\n- Metadata says: only task name \"PROBE\" with no stimulus description.\n- Few-shot pattern suggests modality is derived from stimulus channel (e.g., oddball with auditory+visual pre-cues => Multisensory), but here there is no stimulus information.\n- Alignment: conflict in the sense that few-shot encourages mapping when stimulus details exist; here they do not. Therefore, Modality=Unknown.\n\nType:\n- Metadata says: only \"PROBE\" as a task name; no cognitive construct described.\n- Few-shot pattern suggests Type can be Clinical/Intervention when the dataset’s main purpose is clinical/medical (e.g., dementia resting EEG; PD clinical study). But without task/aim description, we cannot reliably pick a cognitive construct.\n- Alignment: weak/insufficient. Clinical context is evident, but the research purpose is not stated. Therefore, Type=Unknown (rather than inferring Clinical/Intervention).","decision_summary":"Pathology top-2:\n1) Surgery (selected): supported by readme quote \"Raw iEEG data + pre and post surgery MRI\" indicating a surgical patient context.\n2) Epilepsy (runner-up): plausible for iEEG+surgery, but NOT explicitly stated anywhere in metadata.\nAlignment status: no conflict with explicit diagnosis (none provided); selected label uses the only explicit clinical recruitment clue (surgery).\nConfidence justification: single explicit quote indicating surgery, but no diagnosis specifics.\n\nModality top-2:\n1) Unknown (selected): no stimulus/sensory channel described; only task label \"PROBE\".\n2) Visual (runner-up): common for cognitive tasks, but not supported by any quote.\nAlignment status: insufficient metadata; cannot infer.\nConfidence justification: absence of any modality-defining quotes.\n\nType top-2:\n1) Unknown (selected): task purpose/construct not described.\n2) Clinical/Intervention (runner-up): invasive iEEG with surgery suggests clinical context, but metadata does not state clinical study aims or intervention focus.\nAlignment status: insufficient metadata to map to a construct.\nConfidence justification: no explicit task description/aim quotes beyond the existence of surgery-related imaging."}},"nemar_citation_count":1,"computed_title":"PROBE iEEG","nchans_counts":[],"sfreq_counts":[],"stats_computed_at":"2026-04-22T23:16:00.221889+00:00","total_duration_s":null,"author_year":"DOMENECH2020","size_human":"11.0 GB","canonical_name":null}}