Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning

<p dir="ltr">Mutations in isocitrate dehydrogenase genes IDH1 and IDH2 are frequently found in diffuse and anaplastic astrocytic and oligodendroglial tumours as well as in secondary glioblastomas. As IDH is a very important prognostic, diagnostic and therapeutic biomarker for glioma,...

Full description

Saved in:
Bibliographic Details
Main Author: Sidong Liu (1901311) (author)
Other Authors: Zubair Shah (231886) (author), Aydin Sav (2428) (author), Carlo Russo (7259) (author), Shlomo Berkovsky (18622984) (author), Yi Qian (466487) (author), Enrico Coiera (64865) (author), Antonio Di Ieva (14943) (author)
Published: 2020
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1864513513396895744
author Sidong Liu (1901311)
author2 Zubair Shah (231886)
Aydin Sav (2428)
Carlo Russo (7259)
Shlomo Berkovsky (18622984)
Yi Qian (466487)
Enrico Coiera (64865)
Antonio Di Ieva (14943)
author2_role author
author
author
author
author
author
author
author_facet Sidong Liu (1901311)
Zubair Shah (231886)
Aydin Sav (2428)
Carlo Russo (7259)
Shlomo Berkovsky (18622984)
Yi Qian (466487)
Enrico Coiera (64865)
Antonio Di Ieva (14943)
author_role author
dc.creator.none.fl_str_mv Sidong Liu (1901311)
Zubair Shah (231886)
Aydin Sav (2428)
Carlo Russo (7259)
Shlomo Berkovsky (18622984)
Yi Qian (466487)
Enrico Coiera (64865)
Antonio Di Ieva (14943)
dc.date.none.fl_str_mv 2020-05-07T12:00:00Z
dc.identifier.none.fl_str_mv 10.1038/s41598-020-64588-y
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Isocitrate_dehydrogenase_IDH_status_prediction_in_histopathology_images_of_gliomas_using_deep_learning/25911478
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biomedical and clinical sciences
Clinical sciences
Oncology and carcinogenesis
Engineering
Biomedical engineering
Information and computing sciences
Machine learning
Adult
Deep Learning
Glioma
Humans
Isocitrate Dehydrogenase
Magnetic Resonance Imaging
Middle Aged
Mutation
Prognosis
dc.title.none.fl_str_mv Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Mutations in isocitrate dehydrogenase genes IDH1 and IDH2 are frequently found in diffuse and anaplastic astrocytic and oligodendroglial tumours as well as in secondary glioblastomas. As IDH is a very important prognostic, diagnostic and therapeutic biomarker for glioma, it is of paramount importance to determine its mutational status. The haematoxylin and eosin (H&E) staining is a valuable tool in precision oncology as it guides histopathology-based diagnosis and proceeding patient’s treatment. However, H&E staining alone does not determine the IDH mutational status of a tumour. Deep learning methods applied to MRI data have been demonstrated to be a useful tool in IDH status prediction, however the effectiveness of deep learning on H&E slides in the clinical setting has not been investigated so far. Furthermore, the performance of deep learning methods in medical imaging has been practically limited by small sample sizes currently available. Here we propose a data augmentation method based on the Generative Adversarial Networks (GAN) deep learning methodology, to improve the prediction performance of IDH mutational status using H&E slides. The H&E slides were acquired from 266 grade II-IV glioma patients from a mixture of public and private databases, including 130 IDH-wildtype and 136 IDH-mutant patients. A baseline deep learning model without data augmentation achieved an accuracy of 0.794 (AUC = 0.920). With GAN-based data augmentation, the accuracy of the IDH mutational status prediction was improved to 0.853 (AUC = 0.927) when the 3,000 GAN generated training samples were added to the original training set (24,000 samples). By integrating also patients’ age into the model, the accuracy improved further to 0.882 (AUC = 0.931). Our findings show that deep learning methodology, enhanced by GAN data augmentation, can support physicians in gliomas’ IDH status prediction.</p><h2>Other Information</h2><p dir="ltr">Published in: Scientific Reports<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1038/s41598-020-64588-y" target="_blank">https://dx.doi.org/10.1038/s41598-020-64588-y</a></p>
eu_rights_str_mv openAccess
id Manara2_8daeaa54020ca68c08579dc3c3fc8516
identifier_str_mv 10.1038/s41598-020-64588-y
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/25911478
publishDate 2020
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learningSidong Liu (1901311)Zubair Shah (231886)Aydin Sav (2428)Carlo Russo (7259)Shlomo Berkovsky (18622984)Yi Qian (466487)Enrico Coiera (64865)Antonio Di Ieva (14943)Biomedical and clinical sciencesClinical sciencesOncology and carcinogenesisEngineeringBiomedical engineeringInformation and computing sciencesMachine learningAdultDeep LearningGliomaHumansIsocitrate DehydrogenaseMagnetic Resonance ImagingMiddle AgedMutationPrognosis<p dir="ltr">Mutations in isocitrate dehydrogenase genes IDH1 and IDH2 are frequently found in diffuse and anaplastic astrocytic and oligodendroglial tumours as well as in secondary glioblastomas. As IDH is a very important prognostic, diagnostic and therapeutic biomarker for glioma, it is of paramount importance to determine its mutational status. The haematoxylin and eosin (H&E) staining is a valuable tool in precision oncology as it guides histopathology-based diagnosis and proceeding patient’s treatment. However, H&E staining alone does not determine the IDH mutational status of a tumour. Deep learning methods applied to MRI data have been demonstrated to be a useful tool in IDH status prediction, however the effectiveness of deep learning on H&E slides in the clinical setting has not been investigated so far. Furthermore, the performance of deep learning methods in medical imaging has been practically limited by small sample sizes currently available. Here we propose a data augmentation method based on the Generative Adversarial Networks (GAN) deep learning methodology, to improve the prediction performance of IDH mutational status using H&E slides. The H&E slides were acquired from 266 grade II-IV glioma patients from a mixture of public and private databases, including 130 IDH-wildtype and 136 IDH-mutant patients. A baseline deep learning model without data augmentation achieved an accuracy of 0.794 (AUC = 0.920). With GAN-based data augmentation, the accuracy of the IDH mutational status prediction was improved to 0.853 (AUC = 0.927) when the 3,000 GAN generated training samples were added to the original training set (24,000 samples). By integrating also patients’ age into the model, the accuracy improved further to 0.882 (AUC = 0.931). Our findings show that deep learning methodology, enhanced by GAN data augmentation, can support physicians in gliomas’ IDH status prediction.</p><h2>Other Information</h2><p dir="ltr">Published in: Scientific Reports<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1038/s41598-020-64588-y" target="_blank">https://dx.doi.org/10.1038/s41598-020-64588-y</a></p>2020-05-07T12:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1038/s41598-020-64588-yhttps://figshare.com/articles/journal_contribution/Isocitrate_dehydrogenase_IDH_status_prediction_in_histopathology_images_of_gliomas_using_deep_learning/25911478CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/259114782020-05-07T12:00:00Z
spellingShingle Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
Sidong Liu (1901311)
Biomedical and clinical sciences
Clinical sciences
Oncology and carcinogenesis
Engineering
Biomedical engineering
Information and computing sciences
Machine learning
Adult
Deep Learning
Glioma
Humans
Isocitrate Dehydrogenase
Magnetic Resonance Imaging
Middle Aged
Mutation
Prognosis
status_str publishedVersion
title Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
title_full Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
title_fullStr Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
title_full_unstemmed Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
title_short Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
title_sort Isocitrate dehydrogenase (IDH) status prediction in histopathology images of gliomas using deep learning
topic Biomedical and clinical sciences
Clinical sciences
Oncology and carcinogenesis
Engineering
Biomedical engineering
Information and computing sciences
Machine learning
Adult
Deep Learning
Glioma
Humans
Isocitrate Dehydrogenase
Magnetic Resonance Imaging
Middle Aged
Mutation
Prognosis