five

Facial Action Unit Data for the FAINT project

收藏
DataCite Commons2025-07-03 更新2025-04-09 收录
下载链接:
https://dataverse.nl/citation?persistentId=doi:10.34894/WXPXKW
下载链接
链接失效反馈
官方服务:
资源简介:
The dataset is part of the FAINT project. The study took place at four blood collection centres (BCC Leiden, ‘s-Hertogenbosch, Zwolle, and Utrecht). We invited all blood donors who fit into the following three groups to participate: (1) the control group; with between 5 to 10 previous donations, and no previous experience of vasovagal reactions, (2) the sensitive group; with 5 to 10 previous donations but with the experience of a VVR at the previous donation, and (3) new donor group; first-time donors. <br><br> During data collection, the donors had their regular blood donation procedure consisting of several phases: registration, health check at the donor physician, blood donation, and cafeteria visit. We used a standard RGB video camera to record the face of each blood donor before the blood donation procedure (stage 1: after they completed the questionnaire, stage 2: after registration). At the same time, we asked each blood donor to self-report vasovagal reactions (VVR) experienced at each disting phase of the blood donation. Verbal VVR ratings were noted by the data manager and recorded using a smartphone voice recorder. <br><br> This dataset contains Facial Action Units extracted using OpenFace software (more information https://github.com/TadasBaltrusaitis/OpenFace), along with three groups (1 = control, 2 = sensitive, and 3 = new donors) and binary targets of low and high VVR groups ( 0 = low, 1 = high). <br><br> The study was approved by the Ethics Advisory Board of Sanquin and the Research Ethics and Data Management Committee (REDC#2019172) of the School of Humanities and Digital Sciences at Tilburg University. This study was performed in line with the principles of the Declaration of Helsinki and written informed consent was obtained from all participants. The data collection started on 11th September 2019 and lasted until 30th November 2022.
提供机构:
DataverseNL
创建时间:
2024-04-17
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作