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

Artificial intelligence (AI) in radiology has made great strides in recentyears, but many hurdles remain. Overfitting and lack of generalizabilityrepresent important ongoing challenges hindering accurate and dependableclinical deployment. If AI algorithms can avoid overfitting and achieve truegeneralizability, they can go from the research realm to the forefront ofclinical work. Recently, small data AI approaches such as deep neuroevolution(DNE) have avoided overfitting small training sets. We seek to address bothoverfitting and generalizability by applying DNE to a virtually pooled data setconsisting of images from various institutions. Our use case is classifyingneuroblastoma brain metastases on MRI. Neuroblastoma is well-suited for ourgoals because it is a rare cancer. Hence, studying this pediatric diseaserequires a small data approach. As a tertiary care center, the neuroblastomaimages in our local Picture Archiving and Communication System (PACS) arelargely from outside institutions. These multi-institutional images provide aheterogeneous data set that can simulate real world clinical deployment. As inprior DNE work, we used a small training set, consisting of 30 normal and 30metastasis-containing post-contrast MRI brain scans, with 37% outside images.The testing set was enriched with 83% outside images. DNE converged to atesting set accuracy of 97%. Hence, the algorithm was able to predict imageclass with near-perfect accuracy on a testing set that simulates real-worlddata. Hence, the work described here represents a considerable contributiontoward clinically feasible AI.

Deep neuroevolution for limited, heterogeneous data: proof-of-concept  application to Neuroblastoma brain metastasis using a small virtual pooled  image collection