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

FDG-PET reveals altered brain metabolism in individuals with mild cognitiveimpairment (MCI) and Alzheimer's disease (AD). Some biomarkers derived fromFDG-PET by computer-aided-diagnosis (CAD) technologies have been proved thatthey can accurately diagnosis normal control (NC), MCI, and AD. However, thestudies of identification of early MCI (EMCI) and late MCI (LMCI) with FDG-PETimages are still insufficient. Compared with studies based on fMRI and DTIimages, the researches of the inter-region representation features in FDG-PETimages are insufficient. Moreover, considering the variability in differentindividuals, some hard samples which are very similar with both two classeslimit the classification performance. To tackle these problems, in this paper,we propose a novel bilinear pooling and metric learning network (BMNet), whichcan extract the inter-region representation features and distinguish hardsamples by constructing embedding space. To validate the proposed method, wecollect 998 FDG-PET images from ADNI. Following the common preprocessing steps,90 features are extracted from each FDG-PET image according to the automaticanatomical landmark (AAL) template and then sent into the proposed network.Extensive 5-fold cross-validation experiments are performed for multipletwo-class classifications. Experiments show that most metrics are improvedafter adding the bilinear pooling module and metric losses to the Baselinemodel respectively. Specifically, in the classification task between EMCI andLMCI, the specificity improves 6.38% after adding the triple metric loss, andthe negative predictive value (NPV) improves 3.45% after using the bilinearpooling module.

Bilinear pooling and metric learning network for early Alzheimer's disease identification with FDG-PET images