English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning methods have become very popular for the processing of naturalimages, and were then successfully adapted to the neuroimaging field. As thesemethods are non-transparent, interpretability methods are needed to validatethem and ensure their reliability. Indeed, it has been shown that deep learningmodels may obtain high performance even when using irrelevant features, byexploiting biases in the training set. Such undesirable situations canpotentially be detected by using interpretability methods. Recently, manymethods have been proposed to interpret neural networks. However, this domainis not mature yet. Machine learning users face two major issues when aiming tointerpret their models: which method to choose, and how to assess itsreliability? Here, we aim at providing answers to these questions by presentingthe most common interpretability methods and metrics developed to assess theirreliability, as well as their applications and benchmarks in the neuroimagingcontext. Note that this is not an exhaustive survey: we aimed to focus on thestudies which we found to be the most representative and relevant.

Interpretability of Machine Learning Methods Applied to Neuroimaging