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

In neuroimaging analysis, fMRI can well assess the function changes for braindiseases with no obvious structural lesions. To date, most deep-learning-basedfMRI studies have employed functional connectivity (FC) as the basic featurefor disease classification. However, FC is calculated on time series ofpredefined regions of interest and neglects detailed information contained ineach voxel. Another drawback of using FC is the limited sample size for thetraining of deep models. The low representation ability of FC leads to poorperformance in clinical practice, especially when dealing with multimodalmedical data involving multiple types of visual signals and textual records forbrain diseases. To overcome this bottleneck problem in the fMRI featuremodality, we propose BrainFormer, an end-to-end functional interaction learningmethod for brain disease classification with single fMRI volume. Unliketraditional deep learning methods that construct convolution and transformerson FC, BrainFormer learns the functional interaction from fMRI signals, bymodeling the local cues within each voxel with 3D convolutions and capturingthe global correlations among distant regions with specially designed globalattention mechanisms from shallow layers to deep layers. Meanwhile, BrainFormercan deal with multimodal medical data including fMRI volume, structural MRI, FCfeatures and phenotypic data to achieve more comprehensive brain diseasediagnosis. We evaluate BrainFormer on five independent multi-site datasets onautism, Alzheimer's disease, depression, attention deficit hyperactivitydisorder and headache disorders. The results demonstrate its effectiveness andgeneralizability for multiple brain diseases diagnosis with multimodalfeatures. BrainFormer may promote precision of neuroimaging-based diagnosis inclinical practice and motivate future studies on fMRI analysis.

Multimodal Brain Disease Classification with Functional Interaction Learning from Single fMRI Volume