↓ Skip to main content

Assessing and tuning brain decoders: Cross-validation, caveats, and guidelines

Overview of attention for article published in NeuroImage, January 2017
Altmetric Badge

About this Attention Score

  • In the top 25% of all research outputs scored by Altmetric
  • High Attention Score compared to outputs of the same age (91st percentile)
  • High Attention Score compared to outputs of the same age and source (88th percentile)

Mentioned by

twitter
38 tweeters
peer_reviews
1 peer review site
reddit
1 Redditor

Citations

dimensions_citation
137 Dimensions

Readers on

mendeley
404 Mendeley
citeulike
1 CiteULike
You are seeing a free-to-access but limited selection of the activity Altmetric has collected about this research output. Click here to find out more.
Title
Assessing and tuning brain decoders: Cross-validation, caveats, and guidelines
Published in
NeuroImage, January 2017
DOI 10.1016/j.neuroimage.2016.10.038
Pubmed ID
Authors

Gaël Varoquaux, Pradeep Reddy Raamana, Denis A. Engemann, Andrés Hoyos-Idrobo, Yannick Schwartz, Bertrand Thirion

Abstract

Decoding, ie prediction from brain images or signals, calls for empirical evaluation of its predictive power. Such evaluation is achieved via cross-validation, a method also used to tune decoders' hyper-parameters. This paper is a review on cross-validation procedures for decoding in neuroimaging. It includes a didactic overview of the relevant theoretical considerations. Practical aspects are highlighted with an extensive empirical study of the common decoders in within- and across-subject predictions, on multiple datasets -anatomical and functional MRI and MEG- and simulations. Theory and experiments outline that the popular "leave-one-out" strategy leads to unstable and biased estimates, and a repeated random splits method should be preferred. Experiments outline the large error bars of cross-validation in neuroimaging settings: typical confidence intervals of 10%. Nested cross-validation can tune decoders' parameters while avoiding circularity bias. However we find that it can be favorable to use sane defaults, in particular for non-sparse decoders.

Twitter Demographics

The data shown below were collected from the profiles of 38 tweeters who shared this research output. Click here to find out more about how the information was compiled.

Mendeley readers

The data shown below were compiled from readership statistics for 404 Mendeley readers of this research output. Click here to see the associated Mendeley record.

Geographical breakdown

Country Count As %
Germany 4 <1%
United States 3 <1%
United Kingdom 2 <1%
Australia 1 <1%
France 1 <1%
Malaysia 1 <1%
Chile 1 <1%
Switzerland 1 <1%
Canada 1 <1%
Other 0 0%
Unknown 389 96%

Demographic breakdown

Readers by professional status Count As %
Student > Ph. D. Student 116 29%
Researcher 80 20%
Student > Master 58 14%
Unspecified 42 10%
Student > Doctoral Student 28 7%
Other 80 20%
Readers by discipline Count As %
Neuroscience 93 23%
Unspecified 90 22%
Psychology 68 17%
Computer Science 47 12%
Engineering 36 9%
Other 70 17%

Attention Score in Context

This research output has an Altmetric Attention Score of 20. This is our high-level measure of the quality and quantity of online attention that it has received. This Attention Score, as well as the ranking and number of research outputs shown below, was calculated when the research output was last mentioned on 01 September 2019.
All research outputs
#804,208
of 13,602,595 outputs
Outputs from NeuroImage
#745
of 8,264 outputs
Outputs of similar age
#22,815
of 263,984 outputs
Outputs of similar age from NeuroImage
#21
of 178 outputs
Altmetric has tracked 13,602,595 research outputs across all sources so far. Compared to these this one has done particularly well and is in the 94th percentile: it's in the top 10% of all research outputs ever tracked by Altmetric.
So far Altmetric has tracked 8,264 research outputs from this source. They typically receive more attention than average, with a mean Attention Score of 9.3. This one has done particularly well, scoring higher than 90% of its peers.
Older research outputs will score higher simply because they've had more time to accumulate mentions. To account for age we can compare this Altmetric Attention Score to the 263,984 tracked outputs that were published within six weeks on either side of this one in any source. This one has done particularly well, scoring higher than 91% of its contemporaries.
We're also able to compare this research output to 178 others from the same source and published within six weeks on either side of this one. This one has done well, scoring higher than 88% of its contemporaries.