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

Unsupervised learning can discover various unseen abnormalities, relying onlarge-scale unannotated medical images of healthy subjects. Towards this,unsupervised methods reconstruct a 2D/3D single medical image to detectoutliers either in the learned feature space or from high reconstruction loss.However, without considering continuity between multiple adjacent slices, theycannot directly discriminate diseases composed of the accumulation of subtleanatomical anomalies, such as Alzheimer's Disease (AD). Moreover, no study hasshown how unsupervised anomaly detection is associated with either diseasestages, various (i.e., more than two types of) diseases, or multi-sequenceMagnetic Resonance Imaging (MRI) scans. Therefore, we propose unsupervisedMedical Anomaly Detection Generative Adversarial Network (MADGAN), a noveltwo-step method using GAN-based multiple adjacent brain MRI slicereconstruction to detect brain anomalies at different stages on multi-sequencestructural MRI: (Reconstruction) Wasserstein loss with Gradient Penalty + 100L1 loss-trained on 3 healthy brain axial MRI slices to reconstruct the next 3ones-reconstructs unseen healthy/abnormal scans; (Diagnosis) Average L2 lossper scan discriminates them, comparing the ground truth/reconstructed slices.For training, we use two different datasets composed of 1,133 healthyT1-weighted (T1) and 135 healthy contrast-enhanced T1 (T1c) brain MRI scans fordetecting AD and brain metastases/various diseases, respectively. OurSelf-Attention MADGAN can detect AD on T1 scans at a very early stage, MildCognitive Impairment (MCI), with Area Under the Curve (AUC) 0.727, and AD at alate stage with AUC 0.894, while detecting brain metastases on T1c scans withAUC 0.921.

MADGAN: unsupervised Medical Anomaly Detection GAN using multiple adjacent brain MRI slice reconstruction