Do Deep Neural Networks Outperform Kernel Regression for Functional Connectivity Prediction of Behavior?
release_mvey55qhergadgkrfedisy45zm
by
Tong He, Ru Kong, Avram Holmes, Minh Nguyen, Mert Sabuncu, Simon B. Eickhoff, Danilo Bzdok, Jiashi Feng, B.T. Thomas Yeo
2018
Abstract
There is significant interest in the development and application of deep neural networks (DNNs) to neuroimaging data. A growing literature suggests that DNNs outperform their classical counterparts in a variety of neuroimaging applications, yet there are few direct comparisons of relative utility. Here, we compared the performance of three DNN architectures and a classical machine learning algorithm (kernel regression) in predicting individual phenotypes from whole-brain resting-state functional connectivity (RSFC) patterns. One of the DNNs was a generic fully-connected feedforward neural network, while the other two DNNs were recently published approaches specifically designed to exploit the structure of connectome data. By using a combined sample of almost 10,000 participants from the Human Connectome Project (HCP) and UK Biobank, we showed that the three DNNs do not outperform kernel regression across a wide range of behavioral and demographic measures. Furthermore, the generic feedforward neural network exhibited similar performance to the two state-of-the-art connectome-specific DNNs. We conclude with suggestions on future neuroimaging DNN research, including comparisons with stronger baseline algorithms, minimum sample sizes, transparency of hyperparameter tuning and code availability. Critically, we believe that deep learning remains a promising tool for analyzing neuroimaging data. However, researchers should carefully consider whether and how their applications might benefit from DNNs' advantages over classical alternatives, rather than treat deep learning as a panacea.
In application/xml+jats
format
Archived Files and Locations
application/pdf 3.0 MB
file_ssmhitygxvdf7irnuag7utduaq
|
www.biorxiv.org (repository) web.archive.org (webarchive) |
application/pdf 1.0 MB
file_o54wdgkyhnhefn5qz6kwd7ywmy
|
www.biorxiv.org (repository) web.archive.org (webarchive) |
application/pdf 5.1 MB
file_i2pciy3prnczthf5thfdgz3fzm
|
www.biorxiv.org (repository) web.archive.org (webarchive) |
application/pdf 1.0 MB
file_talut6q3mfbf7ffhrbsha2fmfq
|
www.biorxiv.org (repository) web.archive.org (webarchive) |
application/pdf 1.0 MB
file_mic7ojn3n5c6vew5jxksueriua
|
web.archive.org (webarchive) www.biorxiv.org (web) |
post
Stage
unknown
Date 2018-11-19
10.1101/473603
access all versions, variants, and formats of this works (eg, pre-prints)
Crossref Metadata (via API)
Worldcat
wikidata.org
CORE.ac.uk
Semantic Scholar
Google Scholar