Residual Convolutional Neural Network for the Determination of IDH Status in Low- and High-Grade Gliomas from MR Imaging - PubMed Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Mar 1;24(5):1073-1081.
doi: 10.1158/1078-0432.CCR-17-2236. Epub 2017 Nov 22.

Residual Convolutional Neural Network for the Determination of IDH Status in Low- and High-Grade Gliomas from MR Imaging

Ken Chang  1 , Harrison X Bai  2 , Hao Zhou  3 , Chang Su  4 , Wenya Linda Bi  5 , Ena Agbodza  2 , Vasileios K Kavouridis  6 , Joeky T Senders  6 , Alessandro Boaro  6 , Andrew Beers  1 , Biqi Zhang  7 , Alexandra Capellini  7 , Weihua Liao  8 , Qin Shen  9 , Xuejun Li  10 , Bo Xiao  3 , Jane Cryan  11 , Shakti Ramkissoon  11 , Lori Ramkissoon  11 , Keith Ligon  11 , Patrick Y Wen  12 , Ranjit S Bindra  4 , John Woo  2 , Omar Arnaout  6 , Elizabeth R Gerstner  13 , Paul J Zhang  14 , Bruce R Rosen  1 , Li Yang  15 , Raymond Y Huang  16 , Jayashree Kalpathy-Cramer  17
Affiliations

Residual Convolutional Neural Network for the Determination of IDH Status in Low- and High-Grade Gliomas from MR Imaging

Ken Chang et al. Clin Cancer Res. .

Abstract

Purpose: Isocitrate dehydrogenase (IDH) mutations in glioma patients confer longer survival and may guide treatment decision making. We aimed to predict the IDH status of gliomas from MR imaging by applying a residual convolutional neural network to preoperative radiographic data.Experimental Design: Preoperative imaging was acquired for 201 patients from the Hospital of University of Pennsylvania (HUP), 157 patients from Brigham and Women's Hospital (BWH), and 138 patients from The Cancer Imaging Archive (TCIA) and divided into training, validation, and testing sets. We trained a residual convolutional neural network for each MR sequence (FLAIR, T2, T1 precontrast, and T1 postcontrast) and built a predictive model from the outputs. To increase the size of the training set and prevent overfitting, we augmented the training set images by introducing random rotations, translations, flips, shearing, and zooming.Results: With our neural network model, we achieved IDH prediction accuracies of 82.8% (AUC = 0.90), 83.0% (AUC = 0.93), and 85.7% (AUC = 0.94) within training, validation, and testing sets, respectively. When age at diagnosis was incorporated into the model, the training, validation, and testing accuracies increased to 87.3% (AUC = 0.93), 87.6% (AUC = 0.95), and 89.1% (AUC = 0.95), respectively.Conclusions: We developed a deep learning technique to noninvasively predict IDH genotype in grade II-IV glioma using conventional MR imaging using a multi-institutional data set. Clin Cancer Res; 24(5); 1073-81. ©2017 AACR.

PubMed Disclaimer

Conflict of interest statement

The authors declare no potential conflicts of interest

Figures

Figure 1
Figure 1
(A) Image pre-processing steps in our proposed approach. (B) A modified 34-layer residual neural network architecture was used to predict IDH status. (C) Displays the learning rate schedule. The learning rate was set to .0001 and stepped down to .25 of its value when there is no improvement in the validation loss for 20 consecutive epochs.
Figure 2
Figure 2
The training heuristics tested include a (A) single combined network, (B) dimensional networks, and (C) sequence networks. In the single combined network training heuristic, all sequences and dimensions were inputted into a single network. In the dimensional networks training heuristic, a separate network was trained for each dimension. In the sequence networks training heuristics, a separate network was trained for each MR sequence.
Figure 3
Figure 3
ROC curves for training, validation, and testing sets from training on three patient cohorts for (A) age only, (B) combining sequence networks, and (C) combining sequence networks + age. The testing set AUC for combing sequence networks + age was 0.95.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Cancer Genome Atlas Research Network. Brat DJ, Verhaak RGW, Aldape KD, Yung WKA, Salama SR, et al. Comprehensive, Integrative Genomic Analysis of Diffuse Lower-Grade Gliomas. N Engl J Med [Internet] 2015;372:2481–98. cited 2017 Feb 28. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26061751. - PMC - PubMed
    1. Schomas DA, Laack NNI, Rao RD, Meyer FB, Shaw EG, O’Neill BP, et al. Intracranial low-grade gliomas in adults: 30-year experience with long-term follow-up at Mayo Clinic. Neuro Oncol [Internet] 2009;11:437–45. cited 2017 Feb 28. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19018039. - PMC - PubMed
    1. Pouratian N, Schiff D. Management of Low-Grade Glioma. Curr Neurol Neurosci Rep [Internet] 2010;10:224–31. cited 2017 Apr 7. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20425038. - PMC - PubMed
    1. Venneti S, Huse JT. The Evolving Molecular Genetics of Low-grade Glioma. Adv Anat Pathol [Internet] 2015;22:94–101. cited 2017 Apr 7. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25664944. - PMC - PubMed
    1. Ostrom QT, Gittleman H, Fulop J, Liu M, Blanda R, Kromer C, et al. CBTRUS Statistical Report: Primary Brain and Central Nervous System Tumors Diagnosed in the United States in 2008–2012. Neuro Oncol [Internet] 2015;17(Suppl 4):iv1–iv62. cited 2015 Nov 7. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26511214. - PMC - PubMed

Publication types

MeSH terms