Skip to main content
Log in

Survival-relevant high-risk subregion identification for glioblastoma patients: the MRI-based multiple instance learning approach

  • Neuro
  • Published:
European Radiology Aims and scope Submit manuscript

Abstract

Objectives

Given the glioblastoma (GBM) heterogeneity, survival-relevant high-risk subregions may exist and facilitate prognosis. The study aimed to identify the high-risk subregions on MRI, and to evaluate their survival stratification performance.

Methods

The gross tumor regions (GTRs) were delineated on the normalized MRI of 104 GBM patients. The signal intensity of voxels from 104 GTRs was pooled as global intensity vector, and K-means clustering was performed on it to find the optimal global clusters. Subregions were generated by assigning back voxels that belonged to each global cluster. Finally, a multiple instance learning (MIL) model was built and validated using radiomics features from each subregion. In this process, subregions predicted as positive would be treated as high-risk subregions, and patients with high-risk subregions inside the GTR would be predicted as having short-term survival.

Results

After K-means clustering, three global clusters were fixed and 294 subregions of 104 patients were generated. Then, the subregion-level MIL model was trained and tested by 200 (71 patients) and 94 subregions (33 patients). The accuracy, sensitivity, and specificity for survival stratification were 87.88%, 85.71%, and 89.47%. Furthermore, 41 high-risk subregions were correctly predicted from patients with short-term survival, in which the median overlap rate of non-enhancing component was 60%.

Conclusion

The stratification performance of high-risk subregions identified by the MIL model was higher than the GTR. The non-enhancing area on MRI was the most important component in high-risk subregions. The MIL approach provides a new perspective on the clinical challenges of glioma with coarse-grained labeling.

Key Points

• The performance of high-risk subregions was more promising than the GTR for OS stratification.

• The non-enhancing component was the most important in the high-risk subregions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Abbreviations

2D:

Two-dimensional

3D:

Three-dimensional

CET:

The contrast-enhanced tumor

FLAIR:

Fluid-attenuated inversion recovery

GBM:

Glioblastoma

GLCM:

The co-occurrence matrix

GLRLM:

The gray-level run-length texture matrix

GLSZM:

The gray-level size-zone matrix

GTR:

The gross tumor regions

ICC:

Intraclass correlation coefficient

KPS:

Karnofsky Performance Status

MIL:

Multiple instance learning

MRI:

Magnetic resonance imaging

nCET:

The non-contrast-enhanced tumor

NEC:

The necrosis tumor

OS:

Overall survival

TCGA :

The Cancer Genome Atlas

TCIA :

The Cancer Imaging Archive

TE:

Echo time

TR:

Repetition time

References

  1. Van Meir EG, Hadjipanayis CG, Norden AD, Shu H-K, Wen PY, Olson JJ (2010) Exciting new advances in neuro-oncology: the avenue to a cure for malignant glioma. CA Cancer J Clin 60:166–193

    Article  Google Scholar 

  2. Stupp R, Mason WP, van den Bent MJ et al (2005) Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352:987–996

    Article  CAS  Google Scholar 

  3. Hu LS, Ning S, Eschbacher JM et al (2017) Radiogenomics to characterize regional genetic heterogeneity in glioblastoma. Neuro Oncol 19:128–137

    Article  CAS  Google Scholar 

  4. Zhou M, Scott J, Chaudhury B et al (2017) Radiomics in brain tumor: image assessment, quantitative feature descriptors, and machine-learning approaches. AJNR Am J Neuroradiol 39:208-216.

  5. Chaddad A, Kucharczyk MJ, Daniel P et al (2019) Radiomics in glioblastoma: current status and challenges facing clinical implementation. Front Oncol 9:374

    Article  Google Scholar 

  6. Liu Y, Zhang X, Feng N et al (2018) The effect of glioblastoma heterogeneity on survival stratification: a multimodal MR imaging texture analysis. Acta Radiol 59:1239–1246

    Article  Google Scholar 

  7. Zhang X, Lu H, Tian Q et al (2019) A radiomics nomogram based on multiparametric MRI might stratify glioblastoma patients according to survival. Eur Radiol 29:5528–5538

    Article  Google Scholar 

  8. Liu Y, Xu X, Yin L, Zhang X, Li L, Lu H (2017) Relationship between glioblastoma heterogeneity and survival time: an MR imaging texture analysis. AJNR Am J Neuroradiol 38:1695–1701

    Article  CAS  Google Scholar 

  9. Cui Y, Ren S, Tha KK, Wu J, Shirato H, Li R (2017) Volume of high-risk intratumoral subregions at multi-parametric MR imaging predicts overall survival and complements molecular analysis of glioblastoma. Eur Radiol 27:3583–3592

    Article  Google Scholar 

  10. Cui Y, Tha KK, Terasaka S et al (2016) Prognostic imaging biomarkers in glioblastoma: development and independent validation on the basis of multiregion and quantitative analysis of MR images. Radiology 278:546–553

    Article  Google Scholar 

  11. Liu S, Wang Y, Xu K et al (2017) Relationship between necrotic patterns in glioblastoma and patient survival: fractal dimension and lacunarity analyses using magnetic resonance imaging. Sci Rep 7:8302

    Article  Google Scholar 

  12. Kickingereder P, Burth S, Wick A et al (2016) Radiomic profiling of glioblastoma: identifying an imaging predictor of patient survival with improved performance over established clinical and radiologic risk models. Radiology 280:880–889

    Article  Google Scholar 

  13. Wang K, Wang Y, Fan X et al (2016) Radiological features combined with IDH1 status for predicting the survival outcome of glioblastoma patients. Neuro Oncol 18:589–597

    Article  Google Scholar 

  14. Kickingereder P, Neuberger U, Bonekamp D et al (2017) Radiomic subtyping improves disease stratification beyond key molecular, clinical and standard imaging characteristics in patients with glioblastoma. Neuro Oncol 20:848–857

    Article  Google Scholar 

  15. Ellingson BM, Harris RJ, Woodworth DC et al (2017) Baseline pretreatment contrast enhancing tumor volume including central necrosis is a prognostic factor in recurrent glioblastoma: evidence from single and multicenter trials. Neuro Oncol 19:89–98

    Article  CAS  Google Scholar 

  16. Gerlinger M, Rowan AJ, Horswell S et al (2012) Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 366:883–892

    Article  CAS  Google Scholar 

  17. Sottoriva A, Spiteri I, Piccirillo SG et al (2013) Intratumor heterogeneity in human glioblastoma reflects cancer evolutionary dynamics. Proc Natl Acad Sci U S A 110:4009–4014

    Article  CAS  Google Scholar 

  18. Jain R, Poisson LM, Gutman D et al (2014) Outcome prediction in patients with glioblastoma by using imaging, clinical, and genomic biomarkers: focus on the nonenhancing component of the tumor. Radiology 272:484–493

    Article  Google Scholar 

  19. Beig N, Patel J, Prasanna P et al (2018) Radiogenomic analysis of hypoxia pathway is predictive of overall survival in glioblastoma. Sci Rep 8:7

    Article  Google Scholar 

  20. Boonzaier NR, Larkin TJ, Matys T, van der Hoorn A, Yan JL, Price SJ (2017) Multiparametric MR imaging of diffusion and perfusion in contrast-enhancing and nonenhancing components in patients with glioblastoma. Radiology 284:180–190

    Article  Google Scholar 

  21. Quellec G, Cazuguel G, Cochener B, Lamard M (2017) Multiple-instance learning for medical image and video analysis. IEEE Rev Biomed Eng 10:213–234

    Article  Google Scholar 

  22. Cheplygina V, de Bruijne M, Pluim JPW (2019) Not-so-supervised: a survey of semi-supervised, multi-instance, and transfer learning in medical image analysis. Med Image Anal 54:280–296

    Article  Google Scholar 

  23. Kanungo T, Mount DM, Netanyahu NS, Piatko CD, Silverman R, Wu AY (2002) An efficient k-means clustering algorithm: analysis and implementation. IEEE T Pattern Anal 24:881–892

    Article  Google Scholar 

  24. Wu S, Zheng J, Li Y et al (2017) A radiomics nomogram for the preoperative prediction of lymph node metastasis in bladder Cancer. Clin Cancer Res 23:6904–6911

    Article  CAS  Google Scholar 

  25. Gevaert O, Mitchell LA, Achrol AS et al (2014) Glioblastoma multiforme: exploratory radiogenomic analysis by using quantitative image features. Radiology 273:168–174

    Article  Google Scholar 

  26. Wu G, Shi Z, Chen Y et al (2019) A sparse representation-based radiomics for outcome prediction of higher grade gliomas. Med Phys 46:250–261

    Article  Google Scholar 

  27. Bakas S, Akbari H, Sotiras A et al (2017) Advancing the Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features. Sci Data 4:170117

    Article  Google Scholar 

Download references

Funding

This study has received funding by the National Nature Science Foundation of China (No. 81701658 to Zhang Xi, 81871424 to Liu Yang, and 81801655 to Tian Qiang) and the Major Project of Shaanxi Province (No. 2020JZ-28 to Liu Yang).

Author information

Authors and Affiliations

Authors

Corresponding authors

Correspondence to Xiaowei He or Yang Liu.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Yang Liu.

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, Xiaopan Xu, has significant statistical expertise.

Informed consent

Written informed consent was not required for this study because all the patient data in TCGA was deidentified.

Ethical approval

Institutional Review Board approval was not required because all the data used in this study were selected from the Cancer Genome Atlas (TCIA). After ethical review by NIH, the TCIA is freely available for the scientific research. Followed by the instructions of TCIA, we have referred related articles about TCIA.

Study subjects or cohorts overlap

Some study subjects or cohorts have been previously reported in AJNR (Am J Neuroradiol. 2017. https://doi.org/10.3174/ajnr.A5279).

Methodology

• retrospective

• diagnostic or prognostic study

• multicenter study

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

ESM 1

(DOCX 52 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, X., Lu, D., Gao, P. et al. Survival-relevant high-risk subregion identification for glioblastoma patients: the MRI-based multiple instance learning approach. Eur Radiol 30, 5602–5610 (2020). https://doi.org/10.1007/s00330-020-06912-8

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-020-06912-8

Keywords

Navigation