---
title: CHB-MIT Scalp EEG Dataset Overview
url: https://www.emergentmind.com/topics/chb-mit-scalp-eeg-dataset
type: topic
---

# CHB-MIT Scalp EEG Dataset Overview

The CHB-MIT Scalp EEG Dataset is a publicly available, extensively annotated resource of long-term, multi-channel scalp electroencephalogram (EEG) recordings from pediatric epilepsy patients collected at Boston Children's Hospital and the Massachusetts Institute of Technology. Designed to serve as a reference for benchmarking seizure detection, prediction, and EEG decoding algorithms, the dataset has become a de facto standard in EEG machine learning and signal processing research, supporting a broad spectrum of methodological advances across clinical neuroengineering, computational neuroscience, and bio-signal analytics.

## 1. Dataset Composition and Subject Cohort

The CHB-MIT Scalp EEG Dataset consists of continuous scalp EEG recordings from 23 pediatric patients (denoted CHB01–CHB23) with medically intractable epilepsy [2306.12292, 2601.12279]. The age range spans 1.5 to 19 years [2306.12292], with each subject undergoing multi-day inpatient video-EEG monitoring as part of pre-surgical clinical evaluation. Aggregate recording duration totals approximately 844–900 hours, with each patient contributing between 9 and 42 EDF (European Data Format) files, typically 1 hour per file [2306.12292]. The dataset includes around 160–200 annotated clinical seizures (median per subject ≈ 8, range 3–40) [2510.13815]. All subjects were on anti-seizure medication during data acquisition [2306.12292].

Recordings were obtained with 23 electrodes placed according to the international 10–20 system in a bipolar montage. In certain files, additional electrodes or vagal-nerve stimulation markers are included, but the canonical setup is 23-channel bipolar [2306.12292, 2601.12279]. Universal technical parameters are a sampling rate of 256 Hz and 16-bit resolution [2306.12292, 2601.12279, 2406.19189].

## 2. Seizure Annotation and Phenotyping

Expert epileptologists reviewed the continuous EEG to annotate seizure onset and offset times, constituting the ground truth for all downstream analyses [1404.0404]. Annotations are provided as text files alongside EDFs, specifying the clock times of each electrographic event [2306.12292]. The dataset contains pre-ictal, ictal, and inter-ictal states. Pre-ictal epochs are frequently operationalized as fixed windows immediately preceding seizure onset (typical intervals: 10–30 min for prediction tasks, 10 s for onset localization), with inter-ictal epochs sampled from seizure-free periods well-separated from ictal episodes (e.g., >4 h buffer) [2601.12279, 1404.0404]. Statistical summaries—such as average seizure duration or time-of-day distributions—are not systematically included in the public release [2601.12279].

## 3. Data Preprocessing Pipelines

Preprocessing workflows are highly task-dependent but consistently leverage and build upon the canonical EDF files. Common steps include:

- **Band-pass filtering:** Ranges vary from 0.5–70 Hz [1404.0404] to 1–70 Hz [2510.13815]; notch filtering is universally applied at 60 Hz and sometimes at harmonics (117–123 Hz) to suppress line noise [2601.12279, 1404.0404].
- **High-pass filtering:** To remove baseline drift, with cutoffs at 1 Hz in seizure prediction contexts [2601.12279].
- **Artifact correction:** Ranges from none (for streamlined, hardware-oriented workflows) [2106.08008] to comprehensive pipelines including manual rejection, independent component analysis (ICA), and removal of drowsy/contaminated segments in biomarker discovery [2510.13815].
- **Normalization:** Channelwise z-score normalization is often employed, using the mean and standard deviation of each channel [2601.12279].
- **Epoching:** Non-overlapping windows of fixed length are extracted for analysis, e.g., 1 s [2310.18767], 5 s [2601.12279], 8 s [2106.08008, 2406.19189], 10 s [1404.0404].
- **Feature extraction:** Signals are either used directly (deep learning), decomposed via Discrete Wavelet Transform (DWT) [2106.08008], or reduced to spectral/spatial features (bandpower, connectivity metrics, line-length, etc.) [2510.13815, 2310.18767].

Preprocessing procedures are summarized in the following table:

| Paper (arXiv ID)    | Band/Notch Filter        | Artifact Handling         | Normalization        | Epoch Length        |
|---------------------|-------------------------|--------------------------|----------------------|---------------------|
| 2601.12279          | 1 Hz HPF, 57–63,117–123 Hz notch | None specified              | z-score (per channel) | 5 s                  |
| 2406.19189          | 0.5–50 Hz band (5th order)     | None, robust to artifacts      | None                 | 8 s                  |
| 2510.13815          | 1–70 Hz band, 60 Hz notch      | Manual+ICA+bad-channel        | None                 | ~6 min blocks        |
| 2310.18767          | None specified                 | None, features only           | quantile (per feature)| 1 s                  |
| 2106.08008          | Implicit in DWT (no explicit)  | None, DWT only                | None                 | 2–8 s                |
| 1404.0404           | 0.5–70 Hz band, 59–61 Hz notch | None                         | None                 | 10 s                 |

## 4. Task-Specific Dataset Construction

Common tasks and their cohort definitions:

- **Seizure prediction:** Preictal versus interictal classification. Preictal is typically defined as 5 s epochs within 30 min before onset [2601.12279] or 10 s immediately before onset [1404.0404]. Interictal epochs are taken from periods ≥4 h separate from any seizure, with postictal buffers (e.g., 30 s) enforced [2601.12279]. Ictal states are often excluded from training in prediction paradigms.
- **Seizure detection:** Detection/classification of seizure segments; positive windows are any epochs overlapping with an annotated seizure [2406.19189, 2106.08008]. Extreme class imbalance is managed variously, e.g., via weighted sampling or majority-vote smoothing [2406.19189, 2106.08008].
- **Biomarker extraction:** Selection of interictal, artifact-free, awake, and eyes-open blocks for electrophysiological severity indexing such as DS-Qi [2510.13815].

Dataset splits may be within-subject (e.g., random 70/30 train/test partition) [2601.12279], leave-one-recording-out or nested cross-validation [2406.19189], or per-patient model evaluation protocols [2310.18767].

## 5. Derivative and Balanced Releases

To facilitate reproducible machine-learning (ML) benchmarking, derivative distributions have been released:

- **IEEE DataPort pre-processed version:** Balanced, fixed-length preictal and ictal segments (4096 s per patient, 23 channels), no additional filtering or artifact removal [2306.12292].
- **Meta-EEG:** Fixed 10 s pre-ictal, ictal, 10 s post-ictal, and non-seizure fragments for ML, not yet open-sourced [2306.12292].
- **PhysioNet:** Original EDFs with annotation files for customizable preprocessing and segmentation workflows [2306.12292].

## 6. Applications and Algorithmic Benchmarks

The dataset has catalyzed algorithmic development and benchmarking in:

- **EEG decoding:** Hierarchical convolutional fusion transformers (HCFT), achieving per-patient sensitivity of 99.10%, specificity of 98.82%, and 0.0236 FP/h [2601.12279].
- **Deep learning segmentation:** BERT-based foundation models fine-tuned to CHB-MIT, with LOOCV protocols yielding sensitivity and false-positive rates suitable for clinical deployment (e.g., 0.23 FP/h at 72.6% sensitivity) [2406.19189].
- **Classical ML pipelines:** Support Vector Machines, Random Forest, Extra Trees, AdaBoost, etc., frequently achieving near-perfect sensitivity and sub-unit FP/h in subject-specific settings, including when constrained to only 4 temporal channels for wearable hardware prototyping [2106.08008].
- **Feature embedding methods:** Periodic embedding modules yielding 100% sensitivity and up to 99.02% specificity (SVM) for 1 s epoch-based detection [2310.18767].
- **Information-theoretic and connectivity analyses:** L-SODA for preictal detection (97.1% sensitivity, mean onset latency 2.8 s) [1404.0404]; DS-Qi for electrophysiological disease severity quantification [2510.13815].

## 7. Research Impact and Limitations

The CHB-MIT database’s influence derives from its open availability, standardized annotation schema, and breadth across pediatric epilepsy phenotypes [2306.12292]. It is the prevailing benchmark for pediatric seizure detection and EEG decoding [2406.19189]. Limitations include high inter- and intra-patient variability, imbalanced class distributions, the absence of high-density or invasive channels, limited metadata on comorbidities and pharmacopathology, and variability in montage (channels range from 18 to 23 in downstream studies) [2601.12279, 2510.13815]. Average seizure duration, etiological homogeneity, and finer quantitative EEG state metadata are incompletely documented—necessitating task-specific, post hoc curation in most research use cases [2601.12279].

The dataset’s lasting value lies in catalyzing robust methodology development across signal processing, clinical BCI, and explainable AI for neurocritical care.

Source: https://www.emergentmind.com/topics/chb-mit-scalp-eeg-dataset