Radiogenomics of lower-grade gliomas: machine learning–based MRI texture analysis for predicting 1p/19q codeletion status
To evaluate the potential value of the machine learning (ML)–based MRI texture analysis for predicting 1p/19q codeletion status of lower-grade gliomas (LGG), using various state-of-the-art ML algorithms.
Materials and methods
For this retrospective study, 107 patients with LGG were included from a public database. Texture features were extracted from conventional T2-weighted and contrast-enhanced T1-weighted MRI images, using LIFEx software. Training and unseen validation splits were created using stratified 10-fold cross-validation technique along with minority over-sampling. Dimension reduction was done using collinearity analysis and feature selection (ReliefF). Classifications were done using adaptive boosting, k-nearest neighbours, naive Bayes, neural network, random forest, stochastic gradient descent, and support vector machine. Friedman test and pairwise post hoc analyses were used for comparison of classification performances based on the area under the curve (AUC).
Overall, the predictive performance of the ML algorithms were statistically significantly different, χ2(6) = 26.7, p < 0.001. There was no statistically significant difference among the performance of the neural network, naive Bayes, support vector machine, random forest, and stochastic gradient descent, adjusted p > 0.05. The mean AUC and accuracy values of these five algorithms ranged from 0.769 to 0.869 and from 80.1 to 84%, respectively. The neural network had the highest mean rank with mean AUC and accuracy values of 0.869 and 83.8%, respectively.
The ML-based MRI texture analysis might be a promising non-invasive technique for predicting the 1p/19q codeletion status of LGGs. Using this technique along with various ML algorithms, more than four-fifths of the LGGs can be correctly classified.
• More than four-fifths of the lower-grade gliomas can be correctly classified with machine learning–based MRI texture analysis. Satisfying classification outcomes are not limited to a single algorithm.
• A few-slice-based volumetric segmentation technique would be a valid approach, providing satisfactory predictive textural information and avoiding excessive segmentation duration in clinical practice.
• Feature selection is sensitive to different patient data set samples so that each sampling leads to the selection of different feature subsets, which needs to be considered in future works.
KeywordsArtificial intelligence Machine learning Radiomics Glioma Mutation
Area under the curve
Grey-level co-occurrence matrix
Grey-level run-length matrix
Grey-level zone length matrix
Magnetic resonance imaging
Neighbourhood grey-level difference matrix
The Cancer Imaging Archive
World Health Organisation
The authors state that this work has not received any funding.
Compliance with ethical standards
The scientific guarantor of this publication is Burak Kocak, MD.
Conflict of interest
The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.
Statistics and biometry
One of the authors (Burak Kocak, MD) has significant statistical expertise.
Written informed consent was not required for this study because all patients included in this study are publicly and freely available for scientific purposes.
Institutional Review Board approval was not required because all patients included in this study are publicly and freely available for scientific purposes.
Study subjects or cohorts overlap
Imaging data of 25 patients were partially used in the authors’ previous work in a completely different context. Previous work has been submitted to another journal.
• Diagnostic or prognostic study
• Based on public data
- 5.Cancer Genome Atlas Research Network, Brat DJ, Verhaak RG et al (2015) Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med 372:2481–2498. https://doi.org/10.1056/NEJMoa1402121
- 9.Ino Y, Betensky RA, Zlatescu MC et al (2001) Molecular subtypes of anaplastic oligodendroglioma: implications for patient management at diagnosis. Clin Cancer Res 7:839–845Google Scholar
- 10.Kaloshi G, Benouaich-Amiel A, Diakite F et al (2007) Temozolomide for low-grade gliomas: predictive impact of 1p/19q loss on response and outcome. Neurology 68:1831–1836. https://doi.org/10.1212/01.wnl.0000262034.26310.a2 CrossRefGoogle Scholar
- 11.Reifenberger J, Reifenberger G, Liu L, James CD, Wechsler W, Collins VP (1994) Molecular genetic analysis of oligodendroglial tumors shows preferential allelic deletions on 19q and 1p. Am J Pathol 145:1175–1190Google Scholar
- 23.Erickson B, Akkus Z, Sedlar J, Korfiatis P (2017) Data from LGG-1p19qDeletion. Cancer Imaging Arch. https://doi.org/10.7937/K9/TCIA.2017.dwehtz9v
- 27.Demšar J, Curk T, Erjavec A et al (2013) Orange: data mining toolbox in Python. J Mach Learn Res 14:2349–2353Google Scholar
- 28.Demšar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30Google Scholar
- 34.Zwanenburg A, Leger S, Vallières M, Löck S (2016) Image biomarker standardisation initiative - feature definitions. arXiv:1612.07003Google Scholar