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

Background and Purpose: Pediatric low-grade glioma (pLGG) is the most commontype of brain tumor in children, and identification of molecular markers forpLGG is crucial for successful treatment planning. Convolutional Neural Network(CNN) models for pLGG subtype identification rely on tumor segmentation. Wehypothesize tumor segmentations are suboptimal and thus, we propose to augmentthe CNN models using tumor location probability in MRI data.  Materials and Methods: Our REB-approved retrospective study included MRIFluid-Attenuated Inversion Recovery (FLAIR) sequences of 143 BRAF fused and 71BRAF V600E mutated tumors. Tumor segmentations (regions of interest (ROIs))were provided by a pediatric neuroradiology fellow and verified by a seniorpediatric neuroradiologist. In each experiment, we randomly split the data intodevelopment and test with an 80/20 ratio. We combined the 3D binary ROI masksfor each class in the development dataset to derive the probability densityfunctions (PDF) of tumor location, and developed three pipelines:location-based, CNN-based, and hybrid.  Results: We repeated the experiment with different model initializations anddata splits 100 times and calculated the Area Under Receiver OperatingCharacteristic Curve (AUC). The location-based classifier achieved an AUC of77.90, 95% confidence interval (CI) (76.76, 79.03). CNN-based classifiersachieved AUC of 86.11, CI (84.96, 87.25), while the tumor-location-guided CNNsoutperformed the formers with an average AUC of 88.64 CI (87.57, 89.72), whichwas statistically significant (Student's t-test p-value 0.0018).  Conclusion: We achieved statistically significant improvements byincorporating tumor location into the CNN models. Our results suggest thatmanually segmented ROIs may not be optimal.

Improving Deep Learning Models for Pediatric Low-Grade Glioma Tumors  Molecular Subtype Identification Using 3D Probability Distributions of Tumor  Location