HIGHLIGHTS
- who: Eduardo Conde-Sousa and colleagues from the I, Instituto de Investigação e Inovação em Saúde, Universidade Do, have published the research: HEROHE Challenge: Predicting HER2 Status in Breast Cancer from Hematoxylin-Eosin Whole-Slide Imaging, in the Journal: (JOURNAL) of 23/05/2022
- what: In this work, a deep convolutional neural_network (CNN) based on residual network (ResNet ) architecture was developed to predict 19 biomarkers, including ER, PgR, and HER2, from tissue morphology. 15 of 24 15 of 24 Six teams achieved, in this subset, AUCs equal to or greater than 0 . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.