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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Whereas MRI produces anatomic information about the brain, functional MRI(fMRI) tells us about neural activity within the brain, including how variousregions communicate with each other. The full chorus of conversations withinthe brain is summarized elegantly in the adjacency matrix. Althoughinformation-rich, adjacency matrices typically provide little in the way ofintuition. Whereas trained radiologists viewing anatomic MRI can readilydistinguish between different kinds of brain cancer, a similar determinationusing adjacency matrices would exceed any expert's grasp. Artificialintelligence (AI) in radiology usually analyzes anatomic imaging, providingassistance to radiologists. For non-intuitive data types such as adjacencymatrices, AI moves beyond the role of helpful assistant, emerging asindispensible. We sought here to show that AI can learn to discern between twoimportant brain tumor types, high-grade glioma (HGG) and low-grade glioma(LGG), based on adjacency matrices. We trained a convolutional neural networks(CNN) with the method of deep neuroevolution (DNE), because of the latter'srecent promising results; DNE has produced remarkably accurate CNNs even whenrelying on small and noisy training sets, or performing nuanced tasks. Aftertraining on just 30 adjacency matrices, our CNN could tell HGG apart from LGGwith perfect testing set accuracy. Saliency maps revealed that the networklearned highly sophisticated and complex features to achieve its success.Hence, we have shown that it is possible for AI to recognize brain tumor typefrom functional connectivity. In future work, we will apply DNE to other noisyand somewhat cryptic forms of medical data, including further explorations withfMRI.

Deep neuroevolution to predict primary brain tumor grade from functional MRI adjacency matrices