Data belonging to Predicting the 1p/19q co-deletion status of presumed low grade glioma with an externally validated machine learning algorithm
Published: 27 August 2019| Version 1 | DOI: 10.17632/rssf5nxxby.1
Contributors:
Sebastian van der Voort, Fatih Incekara, Maarten Wijnenga, Georgios Kapas, Mayke Gardeniers, Joost Schouten, Martijn Starmans, Rishie Nandoe Tewarie, Geert Lycklama, Pim French, Hendrikus Dubbink, Martin van den Bent, Arnaud Vincent, Wiro Niessen, Stefan Klein, Marion SmitsDescription
Data belonging to the 'Predicting the 1p/19q co-deletion status of presumed low grade glioma with an externally validated machine learning algorithm' paper, as publisched in Clinical Cancer Research. When using this data please cite: (Citation follows later). Data includes trained SVM models, image features derived for all patients, labels for all patients, PCE models used to derive feature importance and segmentations made for the LGG-1p19qDeletion dataset from The Cancer Imaging Archive (https://wiki.cancerimagingarchive.net/display/Public/LGG-1p19qDeletion#a888d85b04c640eeaf802e12db2dc8ad)
Files
Steps to reproduce
Feature files and trained models can be used by PREDICT (https://github.com/Svdvoort/PREDICT), or can be loaded using sklearn and hdf5.
Institutions
Erasmus MC Biomedical Imaging Group Rotterdam, Medisch Centrum Haaglanden, Erasmus MC Afdeling Radiologie, Erasmus MC Afdeling Neurologie
Categories
Radiology, Genetics, Machine Learning