Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence

Zeynettin Akkus, Issa Ali, Jiří Sedlář, Jay Paul Agrawal, Ian F. Parney, Caterina Giannini, Bradley J. Erickson

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Several studies have linked codeletion of chromosome arms 1p/19q in low-grade gliomas (LGG) with positive response to treatment and longer progression-free survival. Hence, predicting 1p/19q status is crucial for effective treatment planning of LGG. In this study, we predict the 1p/19q status from MR images using convolutional neural networks (CNN), which could be a non-invasive alternative to surgical biopsy and histopathological analysis. Our method consists of three main steps: image registration, tumor segmentation, and classification of 1p/19q status using CNN. We included a total of 159 LGG with 3 image slices each who had biopsy-proven 1p/19q status (57 non-deleted and 102 codeleted) and preoperative postcontrast-T1 (T1C) and T2 images. We divided our data into training, validation, and test sets. The training data was balanced for equal class probability and was then augmented with iterations of random translational shift, rotation, and horizontal and vertical flips to increase the size of the training set. We shuffled and augmented the training data to counter overfitting in each epoch. Finally, we evaluated several configurations of a multi-scale CNN architecture until training and validation accuracies became consistent. The results of the best performing configuration on the unseen test set were 93.3% (sensitivity), 82.22% (specificity), and 87.7% (accuracy). Multi-scale CNN with their self-learning capability provides promising results for predicting 1p/19q status non-invasively based on T1C and T2 images. Predicting 1p/19q status non-invasively from MR images would allow selecting effective treatment strategies for LGG patients without the need for surgical biopsy.

Original languageEnglish (US)
Pages (from-to)469-476
Number of pages8
JournalJournal of Digital Imaging
Volume30
Issue number4
DOIs
StatePublished - Aug 1 2017

Fingerprint

Artificial Intelligence
Glioma
Biopsy
Neural networks
Image registration
Chromosomes
Network architecture
Disease-Free Survival
Tumors
Learning
Planning
Sensitivity and Specificity
Therapeutics
Neoplasms

All Science Journal Classification (ASJC) codes

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Science Applications

Cite this

Akkus, Z., Ali, I., Sedlář, J., Agrawal, J. P., Parney, I. F., Giannini, C., & Erickson, B. J. (2017). Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence. Journal of Digital Imaging, 30(4), 469-476. https://doi.org/10.1007/s10278-017-9984-3
Akkus, Zeynettin ; Ali, Issa ; Sedlář, Jiří ; Agrawal, Jay Paul ; Parney, Ian F. ; Giannini, Caterina ; Erickson, Bradley J. / Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence. In: Journal of Digital Imaging. 2017 ; Vol. 30, No. 4. pp. 469-476.
@article{cb14c16957254eb6ace6458c8b8f8400,
title = "Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence",
abstract = "Several studies have linked codeletion of chromosome arms 1p/19q in low-grade gliomas (LGG) with positive response to treatment and longer progression-free survival. Hence, predicting 1p/19q status is crucial for effective treatment planning of LGG. In this study, we predict the 1p/19q status from MR images using convolutional neural networks (CNN), which could be a non-invasive alternative to surgical biopsy and histopathological analysis. Our method consists of three main steps: image registration, tumor segmentation, and classification of 1p/19q status using CNN. We included a total of 159 LGG with 3 image slices each who had biopsy-proven 1p/19q status (57 non-deleted and 102 codeleted) and preoperative postcontrast-T1 (T1C) and T2 images. We divided our data into training, validation, and test sets. The training data was balanced for equal class probability and was then augmented with iterations of random translational shift, rotation, and horizontal and vertical flips to increase the size of the training set. We shuffled and augmented the training data to counter overfitting in each epoch. Finally, we evaluated several configurations of a multi-scale CNN architecture until training and validation accuracies became consistent. The results of the best performing configuration on the unseen test set were 93.3{\%} (sensitivity), 82.22{\%} (specificity), and 87.7{\%} (accuracy). Multi-scale CNN with their self-learning capability provides promising results for predicting 1p/19q status non-invasively based on T1C and T2 images. Predicting 1p/19q status non-invasively from MR images would allow selecting effective treatment strategies for LGG patients without the need for surgical biopsy.",
author = "Zeynettin Akkus and Issa Ali and Jiř{\'i} Sedl{\'a}ř and Agrawal, {Jay Paul} and Parney, {Ian F.} and Caterina Giannini and Erickson, {Bradley J.}",
year = "2017",
month = "8",
day = "1",
doi = "10.1007/s10278-017-9984-3",
language = "English (US)",
volume = "30",
pages = "469--476",
journal = "Journal of Digital Imaging",
issn = "0897-1889",
publisher = "Springer New York",
number = "4",

}

Akkus, Z, Ali, I, Sedlář, J, Agrawal, JP, Parney, IF, Giannini, C & Erickson, BJ 2017, 'Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence', Journal of Digital Imaging, vol. 30, no. 4, pp. 469-476. https://doi.org/10.1007/s10278-017-9984-3

Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence. / Akkus, Zeynettin; Ali, Issa; Sedlář, Jiří; Agrawal, Jay Paul; Parney, Ian F.; Giannini, Caterina; Erickson, Bradley J.

In: Journal of Digital Imaging, Vol. 30, No. 4, 01.08.2017, p. 469-476.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence

AU - Akkus, Zeynettin

AU - Ali, Issa

AU - Sedlář, Jiří

AU - Agrawal, Jay Paul

AU - Parney, Ian F.

AU - Giannini, Caterina

AU - Erickson, Bradley J.

PY - 2017/8/1

Y1 - 2017/8/1

N2 - Several studies have linked codeletion of chromosome arms 1p/19q in low-grade gliomas (LGG) with positive response to treatment and longer progression-free survival. Hence, predicting 1p/19q status is crucial for effective treatment planning of LGG. In this study, we predict the 1p/19q status from MR images using convolutional neural networks (CNN), which could be a non-invasive alternative to surgical biopsy and histopathological analysis. Our method consists of three main steps: image registration, tumor segmentation, and classification of 1p/19q status using CNN. We included a total of 159 LGG with 3 image slices each who had biopsy-proven 1p/19q status (57 non-deleted and 102 codeleted) and preoperative postcontrast-T1 (T1C) and T2 images. We divided our data into training, validation, and test sets. The training data was balanced for equal class probability and was then augmented with iterations of random translational shift, rotation, and horizontal and vertical flips to increase the size of the training set. We shuffled and augmented the training data to counter overfitting in each epoch. Finally, we evaluated several configurations of a multi-scale CNN architecture until training and validation accuracies became consistent. The results of the best performing configuration on the unseen test set were 93.3% (sensitivity), 82.22% (specificity), and 87.7% (accuracy). Multi-scale CNN with their self-learning capability provides promising results for predicting 1p/19q status non-invasively based on T1C and T2 images. Predicting 1p/19q status non-invasively from MR images would allow selecting effective treatment strategies for LGG patients without the need for surgical biopsy.

AB - Several studies have linked codeletion of chromosome arms 1p/19q in low-grade gliomas (LGG) with positive response to treatment and longer progression-free survival. Hence, predicting 1p/19q status is crucial for effective treatment planning of LGG. In this study, we predict the 1p/19q status from MR images using convolutional neural networks (CNN), which could be a non-invasive alternative to surgical biopsy and histopathological analysis. Our method consists of three main steps: image registration, tumor segmentation, and classification of 1p/19q status using CNN. We included a total of 159 LGG with 3 image slices each who had biopsy-proven 1p/19q status (57 non-deleted and 102 codeleted) and preoperative postcontrast-T1 (T1C) and T2 images. We divided our data into training, validation, and test sets. The training data was balanced for equal class probability and was then augmented with iterations of random translational shift, rotation, and horizontal and vertical flips to increase the size of the training set. We shuffled and augmented the training data to counter overfitting in each epoch. Finally, we evaluated several configurations of a multi-scale CNN architecture until training and validation accuracies became consistent. The results of the best performing configuration on the unseen test set were 93.3% (sensitivity), 82.22% (specificity), and 87.7% (accuracy). Multi-scale CNN with their self-learning capability provides promising results for predicting 1p/19q status non-invasively based on T1C and T2 images. Predicting 1p/19q status non-invasively from MR images would allow selecting effective treatment strategies for LGG patients without the need for surgical biopsy.

UR - http://www.scopus.com/inward/record.url?scp=85020663461&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020663461&partnerID=8YFLogxK

U2 - 10.1007/s10278-017-9984-3

DO - 10.1007/s10278-017-9984-3

M3 - Article

C2 - 28600641

AN - SCOPUS:85020663461

VL - 30

SP - 469

EP - 476

JO - Journal of Digital Imaging

JF - Journal of Digital Imaging

SN - 0897-1889

IS - 4

ER -