期刊
NEUROIMAGE
卷 241, 期 -, 页码 -出版社
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.neuroimage.2021.118423
关键词
Variational autoencoder; Deep generative model; Unsupervised learning; Latent gradients
资金
- National Institute of Mental Health [R01MH104402]
- University of Michigan
By training a variational autoencoder with rsfMRI data, researchers have been able to untangle the underlying sources of brain cortical activity and connectivity, representing spatiotemporal characteristics and driving changes in cortical networks. The resultant latent variables can be used as a reliable feature for accurate subject identification, even with limited data available. This demonstrates the value of VAE for unsupervised representation learning in resting state fMRI activity.
Resting state functional magnetic resonance imaging (rsfMRI) data exhibits complex but structured patterns. However, the underlying origins are unclear and entangled in rsfMRI data. Here we establish a variational auto encoder, as a generative model trainable with unsupervised learning, to disentangle the unknown sources of rsfMRI activity. After being trained with large data from the Human Connectome Project, the model has learned to represent and generate patterns of cortical activity and connectivity using latent variables. The latent representation and its trajectory represent the spatiotemporal characteristics of rsfMRI activity. The latent variables reflect the principal gradients of the latent trajectory and drive activity changes in cortical networks. Representational geometry captured as covariance or correlation between latent variables, rather than cortical connectivity, can be used as a more reliable feature to accurately identify subjects from a large group, even if only a short period of data is available in each subject. Our results demonstrate that VAE is a valuable addition to existing tools, particularly suited for unsupervised representation learning of resting state fMRI activity.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据