Original Research

Machine learning in preoperative glioma MRI: Survival associations by perfusion-based support vector machine outperforms traditional MRI

Corresponding Author

Kyrre E. Emblem PhD

Department of Radiology and Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA

Intervention Centre, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Address reprint requests to: K.E.E., Massachusetts General Hospital, Building 149, 13th St., Charlestown, MA 02129. E-mail: [email protected]Search for more papers by this author

Paulina Due-Tonnessen MD,

Paulina Due-Tonnessen MD

Department of Radiology, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

John K. Hald MD, PhD,

John K. Hald MD, PhD

Department of Radiology, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

Atle Bjornerud PhD,

Atle Bjornerud PhD

Intervention Centre, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Department of Physics, University of Oslo, Oslo, Norway

Search for more papers by this author

Marco C. Pinho MD,

Marco C. Pinho MD

Department of Radiology and Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA

Department of Radiology, University of Texas Southwestern Medical Center, Dallas, Texas, USA

Search for more papers by this author

David Scheie MD,

David Scheie MD

Department of Pathology, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

Lothar R. Schad PhD,

Lothar R. Schad PhD

Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany

Search for more papers by this author

Torstein R. Meling MD, PhD,

Torstein R. Meling MD, PhD

Department of Neurosurgery, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

Frank G. Zoellner PhD,

Frank G. Zoellner PhD

Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany

Search for more papers by this author

Kyrre E. Emblem PhD,

Corresponding Author

Kyrre E. Emblem PhD

Department of Radiology and Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA

Intervention Centre, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Address reprint requests to: K.E.E., Massachusetts General Hospital, Building 149, 13th St., Charlestown, MA 02129. E-mail: [email protected]Search for more papers by this author

Paulina Due-Tonnessen MD,

Paulina Due-Tonnessen MD

Department of Radiology, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

John K. Hald MD, PhD,

John K. Hald MD, PhD

Department of Radiology, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

Atle Bjornerud PhD,

Atle Bjornerud PhD

Intervention Centre, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Department of Physics, University of Oslo, Oslo, Norway

Search for more papers by this author

Marco C. Pinho MD,

Marco C. Pinho MD

Department of Radiology and Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA

Department of Radiology, University of Texas Southwestern Medical Center, Dallas, Texas, USA

Search for more papers by this author

David Scheie MD,

David Scheie MD

Department of Pathology, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

Lothar R. Schad PhD,

Lothar R. Schad PhD

Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany

Search for more papers by this author

Torstein R. Meling MD, PhD,

Torstein R. Meling MD, PhD

Department of Neurosurgery, Rikshospitalet, Oslo University Hospital, Oslo, Norway

Search for more papers by this author

Frank G. Zoellner PhD,

Frank G. Zoellner PhD

Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany

Search for more papers by this author

First published: 13 November 2013

https://doi.org/10.1002/jmri.24390

Citations: 36

Share a link

Email
Wechat
Bluesky

Abstract

Purpose

To retrospectively evaluate the performance of an automatic support vector machine (SVM) routine in combination with perfusion-based dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) for preoperative survival associations in patients with gliomas and compare our results to traditional MRI.

Materials and Methods

The study was approved by the Ethics Committee and informed consent was signed. Structural, diffusion- and perfusion-weighted MRI was performed at 1.5-T preoperatively in 94 adult patients (49 males, 45 females, 23–82 years; mean 51 years) later diagnosed with a primary glioma. Patients were randomly assigned in training and test datasets and the resulting DSC-based survival associations by SVM were compared to traditional MRI features including contrast-agent enhancement, perfusion- and diffusion-weighted imaging, tumor size, and location. The results were adjusted for age, neurological status, and postoperative factors associated with survival, including surgery and adjuvant therapy.

Results

For 1- (26/33 alive, 11/14 deceased), 2- (15/21, 21/26), 3- (12/16, 27/31) and 4- (12/15, 28/32) year survival associations in the test dataset (47 patients), the SVM routine was the only biomarker to consistently associate with survival (Cox; P < 0.001).

Conclusion

The automatic machine learning routine presented in our study may provide the operator with a reliable instrument for assessing survival in patients with glioma. J. Magn. Reson. Imaging 2014;40:47–54. © 2013 Wiley Periodicals, Inc.

REFERENCES

1Sorensen AG. Perfusion MR imaging: moving forward. Radiology 2008; 249: 416–417.
10.1148/radiol.2492081429
PubMed Web of Science® Google Scholar
2Emblem KE, Nedregaard B, Nome T, et al. Glioma grading by using histogram analysis of blood volume heterogeneity from MR-derived cerebral blood volume maps. Radiology 2008; 247: 808–817.
10.1148/radiol.2473070571
PubMed Web of Science® Google Scholar
3Sorensen AG, Patel S, Harmath C, et al. Comparison of diameter and perimeter methods for tumor volume calculation. J Clin Oncol 2001; 19: 551–557.
10.1200/JCO.2001.19.2.551
CAS PubMed Web of Science® Google Scholar
4Zacharaki EI, Wang S, Chawla S, et al. Classification of brain tumor type and grade using MRI texture and shape in a machine learning scheme. Magn Reson Med 2009; 62: 1609–1618.
10.1002/mrm.22147
PubMed Web of Science® Google Scholar
5Hu X, Wong KK, Young GS, Guo L, Wong ST. Support vector machine multiparametric MRI identification of pseudoprogression from tumor recurrence in patients with resected glioblastoma. J Magn Reson Imaging 2011; 33: 296–305.
10.1002/jmri.22432
PubMed Web of Science® Google Scholar
6Emblem KE, Zoellner FG, Tennoe B, et al. Predictive modeling in glioma grading from MR perfusion images using support vector machines. Magn Reson Med 2008; 60: 945–952.
10.1002/mrm.21736
PubMed Web of Science® Google Scholar
7Law M, Young RJ, Babb JS, 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: 490–498.
10.1148/radiol.2472070898
PubMed Web of Science® Google Scholar
8Chang EF, Clark A, Smith JS, et al. Functional mapping-guided resection of low-grade gliomas in eloquent areas of the brain: improvement of long-term survival. Clinical article. J Neurosurg 2011; 114: 566–573.
10.3171/2010.6.JNS091246
PubMed Web of Science® Google Scholar
9Brasil CG, Ciccarelli O, Altmann DR, et al. Low-grade gliomas: six-month tumor growth predicts patient outcome better than admission tumor volume, relative cerebral blood volume, and apparent diffusion coefficient. Radiology 2009; 253: 505–512.
10.1148/radiol.2532081623
Web of Science® Google Scholar
10Murakami R, Sugahara T, Nakamura H, et al. Malignant supratentorial astrocytoma treated with postoperative radiation therapy: prognostic value of pretreatment quantitative diffusion-weighted MR imaging. Radiology 2007; 243: 493–499.
10.1148/radiol.2432060450
PubMed Web of Science® Google Scholar
11Tozer DJ, Jager HR, Danchaivijitr N, et al. Apparent diffusion coefficient histograms may predict low-grade glioma subtype. NMR Biomed 2007; 20: 49–57.
10.1002/nbm.1091
PubMed Web of Science® Google Scholar
12Young R, Babb J, Law M, Pollack E, Johnson G. Comparison of region-of-interest analysis with three different histogram analysis methods in the determination of perfusion metrics in patients with brain gliomas. J Magn Reson Imaging 2007; 26: 1053–1063.
10.1002/jmri.21064
CAS PubMed Web of Science® Google Scholar
13Park JK, Hodges T, Arko L, et al. Scale to predict survival after surgery for recurrent glioblastoma multiforme. J Clin Oncol 2010; 28: 3838–3843.
10.1200/JCO.2010.30.0582
PubMed Web of Science® Google Scholar
14Zoellner FG, Emblem KE, Schad LR. Support vector machines in DSC-based glioma imaging: suggestions for optimal characterization. Magn Reson Med 2010; 64: 1230–1236.
10.1002/mrm.22495
PubMed Web of Science® Google Scholar
15Boxerman JL, Schmainda KM, Weisskoff RM. Relative cerebral blood volume maps corrected for contrast agent extravasation significantly correlate with glioma tumor grade, whereas uncorrected maps do not. AJNR Am J Neuroradiol 2006; 27: 859–867.
CAS PubMed Web of Science® Google Scholar
16Bjornerud A, Emblem KE. A fully automated method for quantitative cerebral hemodynamic analysis using DSC-MRI. J Cereb Blood Flow Metab 2010; 30: 1066–1078.
10.1038/jcbfm.2010.4
PubMed Web of Science® Google Scholar
17Caseiras GB, Thornton JS, Yousry T, et al. Inclusion or exclusion of intratumoral vessels in relative cerebral blood volume characterization in low-grade gliomas: does it make a difference? AJNR Am J Neuroradiol 2008; 29: 1140–1141.
10.3174/ajnr.A0993
PubMed Web of Science® Google Scholar
18Emblem KE, Due-Tonnessen P, Hald JK, Bjornerud A. Automatic vessel removal in gliomas from dynamic susceptibility contrast imaging. Magn Reson Med 2009; 61: 1210–1217.
10.1002/mrm.21944
PubMed Web of Science® Google Scholar
19Chang CC, Lin CJ. Training nu-support vector classifiers: theory and algorithms. Neural Comput 2001; 13: 2119–2147.
10.1162/089976601750399335
CAS PubMed Web of Science® Google Scholar
20Vapnik VN. Universal learning technology: support vector machines. NEC J Advanced Technol 2005; 2: 137–144.
Google Scholar
21Wen PY, Macdonald DR, Reardon DA, et al. Updated response assessment criteria for high-grade gliomas: response assessment in neuro-oncology working group. J Clin Oncol 2010; 28: 1963–1972.
10.1200/JCO.2009.26.3541
PubMed Web of Science® Google Scholar
22Eickhoff SB, Stephan KE, Mohlberg H, et al. A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data. Neuroimage 2005; 25: 1325–1335.
10.1016/j.neuroimage.2004.12.034
PubMed Web of Science® Google Scholar
23Helseth R, Helseth E, Johannesen TB, et al. Overall survival, prognostic factors, and repeated surgery in a consecutive series of 516 patients with glioblastoma multiforme. Acta Neurol Scand 2010; 122: 159–167.
10.1111/j.1600-0404.2010.01350.x
CAS PubMed Web of Science® Google Scholar
24Zacharaki EI, Morita N, Bhatt P, O'Rourke DM, Melhem ER, Davatzikos C. Survival analysis of patients with high-grade gliomas based on data mining of imaging variables. AJNR Am J Neuroradiol 2012; 33: 1065–1071.
10.3174/ajnr.A2939
CAS PubMed Web of Science® Google Scholar
25Garzon B, Emblem KE, Mouridsen K, et al. Multiparametric analysis of magnetic resonance images for glioma grading and patient survival time prediction. Acta Radiol 2011; 46: 686–696.
Google Scholar
26Danchaivijitr N, Waldman AD, Tozer DJ, et al. Low-grade gliomas: do changes in rCBV measurements at longitudinal perfusion-weighted MR imaging predict malignant transformation? Radiology 2008; 247: 170–178.
10.1148/radiol.2471062089
PubMed Web of Science® Google Scholar
27Law M, Yang S, Wang H, et al. Glioma grading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with conventional MR imaging. AJNR Am J Neuroradiol 2003; 24: 1989–1998.
PubMed Web of Science® Google Scholar
28Galanaud D, Nicoli F, Chinot O, et al. Noninvasive diagnostic assessment of brain tumors using combined in vivo MR imaging and spectroscopy. Magn Reson Med 2006; 55: 1236–1245.
10.1002/mrm.20886
CAS PubMed Web of Science® Google Scholar
29Hollingworth W, Medina LS, Lenkinski RE, et al. A systematic literature review of magnetic resonance spectroscopy for the characterization of brain tumors. AJNR Am J Neuroradiol 2006; 27: 1404–1411.
CAS PubMed Web of Science® Google Scholar
30Patankar TF, Haroon HA, Mills SJ, et al. Is volume transfer coefficient (K (trans)) related to histologic grade in human gliomas? AJNR Am J Neuroradiol 2005; 26: 2455–2465.
PubMed Web of Science® Google Scholar

Citing Literature

Volume40, Issue1

July 2014

Pages 47-54

Machine learning in preoperative glioma MRI: Survival associations by perfusion-based support vector machine outperforms traditional MRI

Abstract

Purpose

Materials and Methods

Results

Conclusion

REFERENCES

Citing Literature

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley

Machine learning in preoperative glioma MRI: Survival associations by perfusion-based support vector machine outperforms traditional MRI

Abstract

Purpose

Materials and Methods

Results

Conclusion

REFERENCES

Citing Literature

References

Related

Information