Glioma Imaging pp 241-249 | Cite as

Radiomics and Machine Learning

  • Julie Ferris
  • Peter D. Chang
  • Daniel S. ChowEmail author


Radiomics is an emerging field that attempts to quantitatively mine medical images for biomarkers including gene expression (imaging genomics or radiogenomics) that have clinical utility. While attempts have been made to visually decode various imaging features on MRIs of gliomas, an artificial intelligence approach is better suited to tease out pixel-level subtleties that may reflect different mutations. Machine learning, a form of artificial intelligence in which a computer learns what to look for without explicit human programming, has shown the most promise in the advancement of radiomics and imaging genomics for glioma characterization. Deep learning frameworks in particular have achieved high sensitivity and specificity in classifying MR images of gliomas by IDH1, 1p19q codeletion, and MGMT promoter methylation status. As these frameworks continue to improve, radiomics and imaging genomics could potentially serve a role in prognosticating outcomes and directing treatments for patients with gliomas.


Radiomics Radiogenomics Imaging genomics Machine learning Deep learning Texture analysis Neural networks Artificial intelligence Glioblastoma Glioma 


  1. 1.
    Belden CJ, et al. Genetics of glioblastoma: a window into its imaging and histopathologic variability. Radiographics. 2011;31(6):1717–40.CrossRefGoogle Scholar
  2. 2.
    Carrillo JA, et al. Relationship between tumor enhancement, edema, IDH1 mutational status, MGMT promoter methylation, and survival in glioblastoma. AJNR Am J Neuroradiol. 2012;33(7):1349–55.CrossRefGoogle Scholar
  3. 3.
    Metellus P, et al. Absence of IDH mutation identifies a novel radiologic and molecular subtype of WHO grade II gliomas with dismal prognosis. Acta Neuropathol. 2010;120(6):719–29.CrossRefGoogle Scholar
  4. 4.
    Paldor I, et al. Frontal glioblastoma multiforme may be biologically distinct from non-frontal and multilobar tumors. J Clin Neurosci. 2016;34:128–32.CrossRefGoogle Scholar
  5. 5.
    Sonoda Y, et al. Association between molecular alterations and tumor location and MRI characteristics in anaplastic gliomas. Brain Tumor Pathol. 2015;32(2):99–104.CrossRefGoogle Scholar
  6. 6.
    Xiong J, et al. Combination of diffusion tensor imaging and conventional MRI correlates with isocitrate dehydrogenase 1/2 mutations but not 1p/19q genotyping in oligodendroglial tumours. Eur Radiol. 2016;26(6):1705–15.CrossRefGoogle Scholar
  7. 7.
    Qi S, et al. Isocitrate dehydrogenase mutation is associated with tumor location and magnetic resonance imaging characteristics in astrocytic neoplasms. Oncol Lett. 2014;7(6):1895–902.CrossRefGoogle Scholar
  8. 8.
    Price SJ, et al. Less invasive phenotype found in Isocitrate dehydrogenase-mutated glioblastomas than in isocitrate dehydrogenase wild-type glioblastomas: a diffusion-tensor imaging study. Radiology. 2017;283(1):215–21.CrossRefGoogle Scholar
  9. 9.
    Kickingereder P, et al. IDH mutation status is associated with a distinct hypoxia/angiogenesis transcriptome signature which is non-invasively predictable with rCBV imaging in human glioma. Sci Rep. 2015;5:16238.CrossRefGoogle Scholar
  10. 10.
    Law M, et al. Gliomas: predicting time to progression or survival with cerebral blood volume measurements at dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging. Radiology. 2008;247(2):490–8.CrossRefGoogle Scholar
  11. 11.
    Nobusawa S, et al. IDH1 mutations as molecular signature and predictive factor of secondary glioblastomas. Clin Cancer Res. 2009;15(19):6002–7.CrossRefGoogle Scholar
  12. 12.
    Yan H, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360(8):765–73.CrossRefGoogle Scholar
  13. 13.
    Johnson DR, et al. Genetically defined oligodendroglioma is characterized by indistinct tumor borders at MRI. AJNR Am J Neuroradiol. 2017;38(4):678–84.CrossRefGoogle Scholar
  14. 14.
    Fellah S, et al. Multimodal MR imaging (diffusion, perfusion, and spectroscopy): is it possible to distinguish oligodendroglial tumor grade and 1p/19q codeletion in the pretherapeutic diagnosis? AJNR Am J Neuroradiol. 2013;34(7):1326–33.CrossRefGoogle Scholar
  15. 15.
    Jenkinson MD, et al. Apparent diffusion coefficients in oligodendroglial tumors characterized by genotype. J Magn Reson Imaging. 2007;26(6):1405–12.CrossRefGoogle Scholar
  16. 16.
    Ducray F, et al. Predictive and prognostic factors for gliomas. Expert Rev Anticancer Ther. 2011;11(5):781–9.CrossRefGoogle Scholar
  17. 17.
    Wiestler B, et al. ATRX loss refines the classification of anaplastic gliomas and identifies a subgroup of IDH mutant astrocytic tumors with better prognosis. Acta Neuropathol. 2013;126(3):443–51.CrossRefGoogle Scholar
  18. 18.
    Drabycz S, et al. An analysis of image texture, tumor location, and MGMT promoter methylation in glioblastoma using magnetic resonance imaging. NeuroImage. 2010;49(2):1398–405.CrossRefGoogle Scholar
  19. 19.
    Moon WJ, et al. Imaging parameters of high grade gliomas in relation to the MGMT promoter methylation status: the CT, diffusion tensor imaging, and perfusion MR imaging. Neuroradiology. 2012;54(6):555–63.CrossRefGoogle Scholar
  20. 20.
    Romano A, et al. Apparent diffusion coefficient obtained by magnetic resonance imaging as a prognostic marker in glioblastomas: correlation with MGMT promoter methylation status. Eur Radiol. 2013;23(2):513–20.CrossRefGoogle Scholar
  21. 21.
    Ellingson BM, et al. Probabilistic radiographic atlas of glioblastoma phenotypes. AJNR Am J Neuroradiol. 2013;34(3):533–40.CrossRefGoogle Scholar
  22. 22.
    Eoli M, et al. Methylation of O6-methylguanine DNA methyltransferase and loss of heterozygosity on 19q and/or 17p are overlapping features of secondary glioblastomas with prolonged survival. Clin Cancer Res. 2007;13(9):2606–13.CrossRefGoogle Scholar
  23. 23.
    Kanas VG, et al. Learning MRI-based classification models for MGMT methylation status prediction in glioblastoma. Comput Methods Prog Biomed. 2017;140:249–57.CrossRefGoogle Scholar
  24. 24.
    Li H, et al. IDH mutation and MGMT promoter methylation are associated with the pseudoprogression and improved prognosis of glioblastoma multiforme patients who have undergone concurrent and adjuvant temozolomide-based chemoradiotherapy. Clin Neurol Neurosurg. 2016;151:31–6.CrossRefGoogle Scholar
  25. 25.
    Gorlia T, et al. Nomograms for predicting survival of patients with newly diagnosed glioblastoma: prognostic factor analysis of EORTC and NCIC trial 26981-22981/CE.3. Lancet Oncol. 2008;9(1):29–38.CrossRefGoogle Scholar
  26. 26.
    Hegi ME, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352(10):997–1003.CrossRefGoogle Scholar
  27. 27.
    Young RJ, et al. Potential role of preoperative conventional MRI including diffusion measurements in assessing epidermal growth factor receptor gene amplification status in patients with glioblastoma. AJNR Am J Neuroradiol. 2013;34(12):2271–7.CrossRefGoogle Scholar
  28. 28.
    Aghi M, et al. Magnetic resonance imaging characteristics predict epidermal growth factor receptor amplification status in glioblastoma. Clin Cancer Res. 2005;11(24 Pt 1):8600–5.CrossRefGoogle Scholar
  29. 29.
    Gupta A, et al. Pretreatment dynamic susceptibility contrast MRI perfusion in glioblastoma: prediction of EGFR gene amplification. Clin Neuroradiol. 2015;25(2):143–50.CrossRefGoogle Scholar
  30. 30.
    Gutman DA, et al. MR imaging predictors of molecular profile and survival: multi-institutional study of the TCGA glioblastoma data set. Radiology. 2013;267(2):560–9.CrossRefGoogle Scholar
  31. 31.
    Yu J, et al. Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade II glioma. Eur Radiol. 2017;27(8):3509–22.CrossRefGoogle Scholar
  32. 32.
    Zhang B, et al. Multimodal MRI features predict isocitrate dehydrogenase genotype in high-grade gliomas. Neuro-Oncology. 2017;19(1):109–17.CrossRefGoogle Scholar
  33. 33.
    Korfiatis P, et al. MRI texture features as biomarkers to predict MGMT methylation status in glioblastomas. Med Phys. 2016;43(6):2835–44.CrossRefGoogle Scholar
  34. 34.
    Levner I, et al. Predicting MGMT methylation status of glioblastomas from MRI texture. Med Image Comput Comput Assist Interv. 2009;12(Pt 2):522–30.PubMedGoogle Scholar
  35. 35.
    Kickingereder P, et al. Radiogenomics of glioblastoma: machine learning-based classification of molecular characteristics by using multiparametric and multiregional MR imaging features. Radiology. 2016;281:161382.CrossRefGoogle Scholar
  36. 36.
    PanY, et al.Brain tumor grading based on Neural Networks and Convolutional Neural Networks. in 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC);2015.Google Scholar
  37. 37.
    Korfiatis P, et al. Residual deep convolutional neural network predicts MGMT methylation status. J Digit Imaging. 2017;30(5):622–8.CrossRefGoogle Scholar
  38. 38.
    Akkus Z, et al. Predicting deletion of chromosomal arms 1p/19q in low-grade gliomas from MR images using machine intelligence. J Digit Imaging. 2017;30(4):469–76.CrossRefGoogle Scholar
  39. 39.
    Chang P, et al. Deep-learning convolutional neural networks accurately classify genetic mutations in gliomas. AJNR Am J Neuroradiol. 2018;39(7):1201–7.CrossRefGoogle Scholar
  40. 40.
    SimonyanK, VedaldiA, ZissermanA.Deep inside convolutional networks: visualising image classification models and saliency maps. CoRR.2013. abs/1312.6034.Google Scholar
  41. 41.
    SelvarajuRR, et al.Grad-CAM: why did you say that? Visual explanations from deep networks via gradient-based localization. CoRR. 2016. abs/1610.02391.Google Scholar
  42. 42.
    Yu J, et al. Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade II glioma. Eur Radiol. 2016;27(8):3509–22.CrossRefGoogle Scholar
  43. 43.
    ZhangX, et al.IDH Mutation Assessment of Glioma Using Texture Features of Multimodal MR Images. In SPIE Medical Imaging; 2017. International Society for Optics and Photonics.Google Scholar
  44. 44.
    Kickingereder P, et al. Radiogenomics of glioblastoma: machine learning-based classification of molecular characteristics by using multiparametric and multiregional MR imaging features. Radiology. 2016;281(3):907–18.CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Julie Ferris
    • 1
  • Peter D. Chang
    • 2
  • Daniel S. Chow
    • 2
    Email author
  1. 1.University of California, IrvineIrvineUSA
  2. 2.Department of RadiologyUniversity of California, IrvineIrvineUSA

Personalised recommendations