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

Accurate Computer-Assisted Diagnosis, relying on large-scale annotatedpathological images, can alleviate the risk of overlooking the diagnosis.Unfortunately, in medical imaging, most available datasets aresmall/fragmented. To tackle this, as a Data Augmentation (DA) method, 3Dconditional Generative Adversarial Networks (GANs) can synthesize desiredrealistic/diverse 3D images as additional training data. However, no 3Dconditional GAN-based DA approach exists for general bounding box-based 3Dobject detection, while it can locate disease areas with physicians' minimumannotation cost, unlike rigorous 3D segmentation. Moreover, since lesions varyin position/size/attenuation, further GAN-based DA performance requiresmultiple conditions. Therefore, we propose 3D Multi-Conditional GAN (MCGAN) togenerate realistic/diverse 32 X 32 X 32 nodules placed naturally on lungComputed Tomography images to boost sensitivity in 3D object detection. OurMCGAN adopts two discriminators for conditioning: the context discriminatorlearns to classify real vs synthetic nodule/surrounding pairs with noisebox-centered surroundings; the nodule discriminator attempts to classify realvs synthetic nodules with size/attenuation conditions. The results show that 3DConvolutional Neural Network-based detection can achieve higher sensitivityunder any nodule size/attenuation at fixed False Positive rates and overcomethe medical data paucity with the MCGAN-generated realistic nodules---evenexpert physicians fail to distinguish them from the real ones in Visual TuringTest.

Synthesizing Diverse Lung Nodules Wherever Massively: 3D Multi-Conditional GAN-based CT Image Augmentation for Object Detection