Journal of Digital Imaging

, Volume 32, Issue 5, pp 766–772 | Cite as

Toward Automatic Detection of Radiation-Induced Cerebral Microbleeds Using a 3D Deep Residual Network

  • Yicheng Chen
  • Javier E. Villanueva-Meyer
  • Melanie A. Morrison
  • Janine M. LupoEmail author


Cerebral microbleeds, which are small focal hemorrhages in the brain that are prevalent in many diseases, are gaining increasing attention due to their potential as surrogate markers of disease burden, clinical outcomes, and delayed effects of therapy. Manual detection is laborious and automatic detection and labeling of these lesions is challenging using traditional algorithms. Inspired by recent successes of deep convolutional neural networks in computer vision, we developed a 3D deep residual network that can distinguish true microbleeds from false positive mimics of a previously developed technique based on traditional algorithms. A dataset of 73 patients with radiation-induced cerebral microbleeds scanned at 7 T with susceptibility-weighted imaging was used to train and evaluate our model. With the resulting network, we maintained 95% of the true microbleeds in 12 test patients and the average number of false positives was reduced by 89%, achieving a detection precision of 71.9%, higher than existing published methods. The likelihood score predicted by the network was also evaluated by comparing to a neuroradiologist’s rating, and good correlation was observed.


Deep learning Susceptibility-weighted imaging Cerebral microbleeds Convolutional neural networks Automated-detection 


Funding Information

This work was supported by the National Institute for Child Health and Human Development of the National Institutes of Health grant R01HD079568 and GE Healthcare.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest with regard to the content of this manuscript.


  1. 1.
    Linn J, Halpin A, Demaerel P, Ruhland J, Giese AD, Dichgans M, van Buchem MA, Bruckmann H, Greenberg SM: Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology 74(17):1346–1350, 2010CrossRefGoogle Scholar
  2. 2.
    Kato H, Izumiyama M, Izumiyama K, Takahashi A, Itoyama Y: Silent cerebral microbleeds on T2*-weighted MRI: correlation with stroke subtype, stroke recurrence, and leukoaraiosis. Stroke 33(6):1536–1540, 2002CrossRefGoogle Scholar
  3. 3.
    Hanyu H, Tanaka Y, Shimizu S, Takasaki M, Abe K: Cerebral microbleeds in Alzheimer’s disease. J Neurol 250(12):1496–1497, 2003CrossRefGoogle Scholar
  4. 4.
    Kinnunen KM, Greenwood R, Powell JH, Leech R, Hawkins PC, Bonnelle V, Patel MC, Counsell SJ, Sharp DJ: White matter damage and cognitive impairment after traumatic brain injury. Brain 134(Pt 2):449–463, 2011CrossRefGoogle Scholar
  5. 5.
    Lupo JM, Chuang CF, Chang SM, Barani IJ, Jimenez B, Hess CP, Nelson SJ. 7-Tesla susceptibilityweighted imaging to assess the effects of radiotherapy on normal-appearing brain in patients with glioma. Int J Radiat Oncol Biol Phys 82(3):e493–500, 2012CrossRefGoogle Scholar
  6. 6.
    Charidimou A, Krishnan A, Werring DJ, Jäger HR: Cerebral microbleeds: a guide to detection and clinical relevance in different disease settings. Neuroradiology 55(6):655–674, 2013CrossRefGoogle Scholar
  7. 7.
    Wahl M, Anwar M, Hess C, Chang SM, Lupo JM: Relationship between radiation dose and microbleed formation in patients with malignant glioma. Int J Radiat Oncol Biol Phys 96(2S):E68, 2016CrossRefGoogle Scholar
  8. 8.
    Kuijf HJ, de Bresser J, Geerlings MI, Conijn MMA, Viergever MA, Biessels GJ, Vincken KL: Efficient detection of cerebral microbleeds on 7.0 T MR images using the radial symmetry transform. Neuroimage 59(3):2266–2273, 2012CrossRefGoogle Scholar
  9. 9.
    van den Heuvel TLA, Ghafoorian M, van der Eerden AW, Goraj BM, Andriessen TMJC, ter Haar Romeny BM, Platel B: Computer aided detection of brain micro-bleeds in traumatic brain injury. In: Medical Imaging 2015: Computer-Aided Diagnosis, Vol. 9414, 2015, p. 94142FGoogle Scholar
  10. 10.
    Barnes SRS, Haacke EM, Ayaz M, Boikov AS, Kirsch W, Kido D: Semiautomated detection of cerebral microbleeds in magnetic resonance images. Magn Reson Imaging 29(6):844–852, 2011CrossRefGoogle Scholar
  11. 11.
    Qi D, Chen H, Lequan Y, Zhao L, Qin J, Wang D, Mok VC, Shi L, Heng P-A: Automatic detection of cerebral microbleeds from MR images via 3D convolutional neural networks. IEEE Trans Med Imaging 35(5):1182–1195, 2016CrossRefGoogle Scholar
  12. 12.
    Krizhevsky A, Sutskever I, Hinton GE: ImageNet classification with deep convolutional neural networks. Commun ACM 60(6):84–90, 2017CrossRefGoogle Scholar
  13. 13.
    Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC, Fei-Fei L: ImageNet large scale visual recognition challenge. Int J Comput Vis 115(3):211–252, 2015CrossRefGoogle Scholar
  14. 14.
    He K, Zhang X, Ren S, Sun J: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778CrossRefGoogle Scholar
  15. 15.
    Huang G, Liu Z, van der Maaten L, Weinberger KQ: Densely connected convolutional networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 2261–2269CrossRefGoogle Scholar
  16. 16.
    Ioffe S, Szegedy C: Batch normalization: accelerating deep network training by reducing internal covariate shift, arXiv preprint arXiv:1502.03167, 2015Google Scholar
  17. 17.
    Bian W, Hess CP, Chang SM, Nelson SJ, Lupo JM: Computer-aided detection of radiation-induced cerebral microbleeds on susceptibility-weighted MR images. Neuroimage Clin 2:282–290, 2013CrossRefGoogle Scholar
  18. 18.
    Bian W, Banerjee S, Kelly DAC, Hess CP, Larson PEZ, Chang SM, Nelson SJ, Lupo JM: Simultaneous imaging of radiation-induced cerebral microbleeds, arteries and veins, using a multiple gradient echo sequence at 7 Tesla. J Magn Reson Imaging 42(2):269–279, 2015CrossRefGoogle Scholar
  19. 19.
    Lupo JM, Banerjee S, Hammond KE, Kelley DAC, Xu D, Chang SM, Vigneron DB, Majumdar S, Nelson SJ: GRAPPA-based susceptibility-weighted imaging of normal volunteers and patients with brain tumor at 7 T. Magn Reson Imaging 27(4):480–488, 2009CrossRefGoogle Scholar
  20. 20.
    Smith SM: Fast robust automated brain extraction. Hum Brain Mapp 17(3):143–155, 2002CrossRefGoogle Scholar
  21. 21.
    Dahl GE, Sainath TN, Hinton GE: Improving deep neural networks for LVCSR using rectified linear units and dropout, Acoustics, Speech and Signal Processing (ICASSP), IEEE International Conference on, 2013, p 8609Google Scholar
  22. 22.
    F. Chollet and Others, Keras, 2015.Google Scholar
  23. 23.
    Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, Corrado GS, Davis A, Dean J, Devin M, Ghemawat S: Tensorflow: Large-scale machine learning on heterogeneous distributed systems, arXiv preprint arXiv:1603.04467, 2016.Google Scholar
  24. 24.
    Kingma DP, Ba J: Adam: a method for stochastic optimization, arXiv preprint arXiv:1412.6980, 2014Google Scholar
  25. 25.
    Simonyan K, Zisserman A: Very deep convolutional networks for large-scale image recognition, arXiv preprint arXiv:1409.1556, 2014Google Scholar
  26. 26.
    Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A: Going deeper with convolutions. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp 1–9Google Scholar
  27. 27.
    Ren S, He K, Girshick R, Sun J: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39(6):1137–1149, 2017CrossRefGoogle Scholar

Copyright information

© Society for Imaging Informatics in Medicine 2018
corrected publication 2018

Authors and Affiliations

  1. 1.UCSF-UC Berkeley Graduate Program in BioengineeringSan FranciscoUSA
  2. 2.Department of Radiology and Biomedical ImagingUniversity of California San FranciscoSan FranciscoUSA

Personalised recommendations