Kragel 270 participant dataset (18 studies x 15 participants) from Kragel et al. 2018 Nature Neuroscience

<br><br>This dataset is an fmri_data object class object created using CANlab tools (canlab.github.io). It contains 270 single-participant fMRI contrast maps across 18 studies (with 15 participants per study). Studies are grouped into three domains: Pain, Cognitive Control, and Negative Emotion, with 9 studies each. Each domain includes 3 subdomains, with 3 studies each. <br>The dataset is from Kragel et al. 2018, Nature Neuroscience. Kragel, P. A., Kano, M., Van Oudenhove, L., Ly, H. G., Dupont, P., Rubio, A., Delon-Martin, C., Bonaz, B. L., Manuck, S. B., Gianaros, P. J., Ceko, M., Losin, R. E. A., Woo, C. W., Nichols, T. E., Wager, T. D. (2018). Generalizable representations of pain, cognitive control, and negative emotion in medial frontal cortex. <i>Nature neuroscience</i>. 21(2): 283-289. PMCID: PMC5801068.https://www.nature.com/articles/s41593-017-0051-7it's also stored on Neurovault.org as collection #504. Using CANlab tools , you could get it using:[files_on_disk, url_on_neurovault, mycollection, myimages] = retrieve_neurovault_collection(504);This file also has metadata that is not on Neurovault, in data_obj.metadata_table. If you save this file somewhere on your Matlab path, you'll be able to reload and reuse the dataset easily.<pre>[test_images, names] = load_image_set('kragel18_alldata', 'noverbose');<br></pre><br>270×6 tableDomain Subdomain imagenames Studynumber Orig_Studynumber StudyCodes_______________ ___________ __________________________ ___________ ________________ {'Pain' } {'Thermal'} {'ThermalPain1' } 1 10 {'Atlas_2010_EXP' }{'Pain' } {'Thermal'} {'ThermalPain1' } 1 10 {'Atlas_2010_EXP' }{'Pain' } {'Thermal'} {'ThermalPain1' } 1 10 {'Atlas_2010_EXP' }{'Pain' } {'Thermal'} {'ThermalPain1' } 1 10 {'Atlas_2010_EXP' }{'Pain' } {'Thermal'} {'ThermalPain1' } 1 10 {'Atlas_2010_EXP' }.... etcNote: Individual participant images are <i>not</i> on the same magnitude scale across studies. For most applications, you will want to normalize the scale (e.g., L2 norm or z-score images).e.g., in CANlab tools:data_obj = rescale(data_obj, 'l2norm_images');A .nii file format image is also included. These images are rescaled by dividing by the L2 norm of each image. <br>

本数据集为基于CANlab工具（canlab.github.io）构建的fMRI_data对象类数据集。其包含来自18项研究的270张单受试者fMRI对比图，每项研究纳入15名受试者。研究被划分为3个领域：疼痛、认知控制与负性情绪，每个领域各包含9项研究；每个领域下设3个子领域，每个子领域对应3项研究。

本数据集源自Kragel等人2018年发表于《自然神经科学》（*Nature Neuroscience*）的研究：
Kragel, P. A., Kano, M., Van Oudenhove, L., Ly, H. G., Dupont, P., Rubio, A., Delon-Martin, C., Bonaz, B. L., Manuck, S. B., Gianaros, P. J., Ceko, M., Losin, R. E. A., Woo, C. W., Nichols, T. E., Wager, T. D. (2018). Generalizable representations of pain, cognitive control, and negative emotion in medial frontal cortex. *Nature Neuroscience*, 21(2): 283-289. PMCID: PMC5801068. https://www.nature.com/articles/s41593-017-0051-7

该数据集同时以收藏集#504的形式存储于Neurovault.org平台。借助CANlab工具，可通过以下Matlab代码获取该数据集：
matlab
[files_on_disk, url_on_neurovault, mycollection, myimages] = retrieve_neurovault_collection(504);

本文件包含Neurovault平台未收录的元数据，存储于`data_obj.metadata_table`中。若将该文件保存至Matlab搜索路径下，即可轻松重载并复用本数据集。亦可通过以下命令加载数据集：
matlab
[test_images, names] = load_image_set('kragel18_alldata', 'noverbose');

本数据集包含一张270行×6列的表格，字段依次为：领域（Domain）、子领域（Subdomain）、图像名称（imagenames）、研究编号（Studynumber）、原始研究编号（Orig_Studynumber）、研究代码（StudyCodes）。示例表格行如下：

Domain Subdomain imagenames Studynumber Orig_Studynumber StudyCodes
{'Pain'} {'Thermal'} {'ThermalPain1'} 1 10 {'Atlas_2010_EXP'}

（注：原文中该示例行重复多次，此处省略冗余内容）

注意：不同研究的单受试者图像的幅值尺度并不统一。在多数应用场景中，需对图像进行尺度归一化处理（例如采用L2范数或z分数标准化）。以CANlab工具为例，可通过以下代码实现：
matlab
data_obj = rescale(data_obj, 'l2norm_images');

本数据集同时包含.nii格式图像，此类图像已通过除以单张图像的L2范数完成归一化。