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

Convolutional neural networks (CNNs) are one of the most popular models ofArtificial Neural Networks (ANN)s in Computer Vision (CV). A variety ofCNN-based structures were developed by researchers to solve problems like imageclassification, object detection, and image similarity measurement. AlthoughCNNs have shown their value in most cases, they still have a downside: theyeasily overfit when there are not enough samples in the dataset. Most medicalimage datasets are examples of such a dataset. Additionally, many datasets alsocontain both designed features and images, but CNNs can only deal with imagesdirectly. This represents a missed opportunity to leverage additionalinformation. For this reason, we propose a new structure of CNN-based model:CompNet, a composite convolutional neural network. This is a specially designedneural network that accepts combinations of images and designed features asinput in order to leverage all available information. The novelty of thisstructure is that it uses learned features from images to weight designedfeatures in order to gain all information from both images and designedfeatures. With the use of this structure on classification tasks, the resultsindicate that our approach has the capability to significantly reduceoverfitting. Furthermore, we also found several similar approaches proposed byother researchers that can combine images and designed features. To makecomparison, we first applied those similar approaches on LIDC and compared theresults with the CompNet results, then we applied our CompNet on the datasetsthat those similar approaches originally used in their works and compared theresults with the results they proposed in their papers. All these comparisonresults showed that our model outperformed those similar approaches onclassification tasks either on LIDC dataset or on their proposed datasets.

CompNet: A Designated Model to Handle Combinations of Images and Designed features