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/

Zendy Plus

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

Zendy Tools

Zendy Open

Cancer histology reveals disease progression and associated molecularprocesses, and contains rich phenotypic information that is predictive ofoutcome. In this paper, we developed a computational approach based on deeplearning to predict the overall survival and molecular subtypes of gliomapatients from microscopic images of tissue biopsies, reflecting measures ofmicrovascular proliferation, mitotic activity, nuclear atypia, and the presenceof necrosis. Whole-slide images from 663 unique patients [IDH: 333IDH-wildtype, 330 IDH-mutants, 1p/19q: 201 1p/19q non-codeleted, 129 1p/19qcodeleted] were obtained from TCGA. Sub-images that were free of artifacts andthat contained viable tumor with descriptive histologic characteristics wereextracted, which were further used for training and testing a deep neuralnetwork. The output layer of the network was configured in two different ways:(i) a final Cox model layer to output a prediction of patient risk, and (ii) afinal layer with sigmoid activation function, and stochastic gradient decentbased optimization with binary cross-entropy loss. Both survival prediction andmolecular subtype classification produced promising results using our model.The c-statistic was estimated to be 0.82 (p-value=4.8x10-5) between the riskscores of the proposed deep learning model and overall survival, whileaccuracies of 88% (area under the curve [AUC]=0.86) were achieved in thedetection of IDH mutational status and 1p/19q codeletion. These findingssuggest that the deep learning techniques can be applied to microscopic imagesfor objective, accurate, and integrated prediction of outcome for gliomapatients. The proposed marker may contribute to (i) stratification of patientsinto clinical trials, (ii) patient selection for targeted therapy, and (iii)personalized treatment planning.

Prediction of overall survival and molecular markers in gliomas via analysis of digital pathology images using deep learning