Electroencephalogram (EEG)

The SEED dataset contains subjects' EEG signals when they were watching films clips. The film clips are carefully selected so as to induce different types of emotion, which are positive, negative, and neutral ones. Source: http://bcmi.sjtu.edu.cn/home/seed/index.html Image Source: http://bcmi.sjtu.edu.cn/home/seed/index.html

An Independent components (IC) dataset containing spatiotemporal measures for over 200,000 ICs from more than 6,000 EEG recordings. Source: ICLabel: An automated electroencephalographic independent component classifier, dataset, and website

This is a gun detection dataset with 51K annotated gun images for gun detection and other 51K cropped gun chip images for gun classification collected from a few different sources.

A multimodal dataset with comprehensive annotations of continuous emotions during naturalistic conversations. The dataset contains multimodal measurements, including audiovisual recordings, EEG, and peripheral physiological signals, acquired with off-the-shelf devices from 16 sessions of approximately 10-minute long paired debates on a social issue. Source: K-EmoCon, a multimodal sensor dataset for continuous emotion recognition in naturalistic conversations

A multimodal database for eye blink detection and attention level estimation. Source: mEBAL: A Multimodal Database for Eye Blink Detection and Attention Level Estimation

The dataset contains a collection of physiological signals (EEG, GSR, PPG) obtained from an experiment of the auditory attention on natural speech. Ethical Approval was acquired for the experiment. Details of the experiment can be found here https://phyaat.github.io/experiment ### Dataset The dataset contain three physiological signals recorded at sampling rate of 128Hz from 25 healthy subjects during the experiment. Electroenceplogram (EEG) signal is recorded using a 14-channel Emotiv Epoc device. Two signal streams of Galvanic Skin Response (GSR) were recorded, instantaneous sample and moving averaged signal. From photoplethysmogram (PPG) sensor (pulse sensor), a raw signal, inter-beat interval (IBI), and pulse rate were recorded. All the signals were properly labeled. - EEG Channels: 'AF3', 'F7', 'F3', 'FC5', 'T7', 'P7', 'O1', 'O2', 'P8', 'T8', 'FC6', 'F4', 'F8', 'AF4' - GSR Signal: Instantaneous and moving averaged signal streams - PPG: PPG (ECG like signal), IBI (Inter Beat Interval ) and BPM (Beats per minute) ### Download the dataset #### Using Python To download the dataset, install phyaat library and download through it. pip install phyaat import phyaat as ph #to download dataset of subject 1 in given path 'dirpath dirPath = ph.download_data(baseDir='../PhyAAt_Data', subject=1,verbose=0,overwrite=False) #to download dataset of all the subjects dirPath = ph.download_data(baseDir='../PhyAAt_Data', subject=-1,verbose=0,overwrite=False) #### Manually If you are using other programming framework such as matlab or R, Download dataset manually from Github repository and extract all the csv files. For more details on downloading and using dataset, check here: Getting Started ### Helper Scripts There are starter scripts and benchmark code to start building models. They are available here - https://phyaat.github.io/modeling/

EEG/fMRI Data from 8 subject doing a simple eyes open/eyes closed task is provided on this webpage. The EEG/fMRI data are six files for each subject, with two basic factors: recording during Helium pump On and Helium pump Off, and recording during MRI scanning and without MRI scanning. In addition 'outside' EEG data is provided, before as well as after the MRI session. There are 30 EEG channels, 1 EOG channel, 1 ECG channel, as well as 6 CWL signals.

This dataset includes time-synchronized multimodal data records of students (learning logs, videos, EEG brainwaves) as they work in various subjects from Squirrel AI Learning System (SAIL) to solve problems of varying difficulty levels. The dataset resources include user records from the learner records store of SAIL, brainwave data collected by EEG headset devices, and video data captured by web cameras while students worked in the SAIL products. Source: MUTLA: A Large-Scale Dataset for Multimodal Teaching and Learning Analytics

Losorelli, Steven, Nguyen, Duc T., Dmochowski, Jacek P., and Kaneshiro, Blair This dataset contains cortical (EEG) and behavioral data collected during natural music listening. Dense-array EEG was recorded from 20 adult participants who each heard a set of 10 full-length songs with electronically produced beats at various tempos. In a separate subsequent listen, each participant tapped to the beat of a 35-second excerpt from each song. Participants also delivered ratings of familiarity and enjoyment for each full-length song during the EEG recording. Finally, the dataset includes basic demographic information about the participants, as well as Matlab scripts to perform the illustrated analyses presented in the paper introducing the dataset (Losorelli et al., 2017). Cleaned and aggregated data are published in Matlab format; raw EEG is published in Matlab format, while raw tapping data are published in .txt format. Stimulus audio is not published, but metadata links are provided.

Please refer to the Zenodo page for a detailed description: https://zenodo.org/records/15665101

Seizures and seizure-like rhythmic and periodic brain activity known as “ictal-interictal-injury continuum” (IIIC) patterns are frequently detected during brain monitoring with electroencephalography (EEG) in patients with epilepsy or critical illness. Prior efforts to automate detection of IIIC patterns have been limited by lack of large well-annotated datasets to train/evaluate algorithms, and there have been only a few attempts to detect IIIC events other than seizures. The IIIC dataset includes 50,697 labeled EEG samples from 2,711 patients’ and 6,095 EEGs that were annotated by physician experts from 18 institutions. These samples were used to train SPaRCNet (Seizures, Periodic and Rhythmic Continuum patterns Deep Neural Network), a computer program that classifies IIIC events with an accuracy matching clinical experts. Prior efforts to automate seizure detection have been limited by lack of large well-annotated datasets to train and evaluate algorithms, and there have been only a few attempts to detect IIIC events other than seizures. To address this gap, we created a set of 50,697 IIIC and non-IIIC events from 2,711 patients’ (6,095 EEGs) and obtained independent annotations from 124 raters