Skip to main content

Advertisement

SpringerLink
Log in

A Comprehensive Study of Data Augmentation Strategies for Prostate Cancer Detection in Diffusion-Weighted MRI Using Convolutional Neural Networks

  • Original Paper
  • Published:
Journal of Digital Imaging Aims and scope Submit manuscript

Abstract

Data augmentation refers to a group of techniques whose goal is to battle limited amount of available data to improve model generalization and push sample distribution toward the true distribution. While different augmentation strategies and their combinations have been investigated for various computer vision tasks in the context of deep learning, a specific work in the domain of medical imaging is rare and to the best of our knowledge, there has been no dedicated work on exploring the effects of various augmentation methods on the performance of deep learning models in prostate cancer detection. In this work, we have statically applied five most frequently used augmentation techniques (random rotation, horizontal flip, vertical flip, random crop, and translation) to prostate diffusion-weighted magnetic resonance imaging training dataset of 217 patients separately and evaluated the effect of each method on the accuracy of prostate cancer detection. The augmentation algorithms were applied independently to each data channel and a shallow as well as a deep convolutional neural network (CNN) was trained on the five augmented sets separately. We used area under receiver operating characteristic (ROC) curve (AUC) to evaluate the performance of the trained CNNs on a separate test set of 95 patients, using a validation set of 102 patients for finetuning. The shallow network outperformed the deep network with the best 2D slice-based AUC of 0.85 obtained by the rotation method.

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

Access this article

Log in via an institution

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15

Similar content being viewed by others

References

  1. Welch HG and Black WC: Overdiagnosis in cancer. J Natl Cancer Inst 102: 605–613, 2010.

    Article  Google Scholar 

  2. Thompson JE, Van Leeuwen PJ, Moses D, Shnier R, Brenner P, Delprado W, Pulbrook M, Böhm M, Haynes AM, Hayen A and Stricker PD: The diagnostic performance of multiparametric magnetic resonance imaging to detect significant prostate cancer. J Urol 195: 1428–1435, 2016.

    Article  CAS  Google Scholar 

  3. Razzak MI, Naz S and Zaib A: Deep learning for medical image processing: Overview, challenges and the future. Lect Notes Comput Vis Biomech 26: 323–350, 2018.

    Article  Google Scholar 

  4. Cao R, Bajgiran AM, Mirak SA, Shakeri S, Zhong X, Enzmann D, Raman S and Sung K: Joint Prostate Cancer Detection and Gleason Score Prediction in mp-MRI via FocalNet. IEEE Trans Med Imaging PP: 1–1, 2019.

  5. Campanella G, Hanna MG, Geneslaw L, Miraflor A, Werneck Krauss Silva V, Busam KJ, Brogi E, Reuter VE, Klimstra DS and Fuchs TJ: Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med 25: 1301–1309, 2019.

    Article  CAS  Google Scholar 

  6. Ishioka J, Matsuoka Y, Uehara S, Yasuda Y, Kijima T, Yoshida S, Yokoyama M, Saito K, Kihara K, Numao N, Kimura T, Kudo K, Kumazawa I and Fujii Y: Computer-aided diagnosis of prostate cancer on magnetic resonance imaging using a convolutional neural network algorithm. BJU Int 122: 411–417, 2018.

    Article  Google Scholar 

  7. Wang X, Yang W, Weinreb J, Han J, Li Q, Kong X, Yan Y, Ke Z, Luo B, Liu T and Wang L: Searching for prostate cancer by fully automated magnetic resonance imaging classification: Deep learning versus non-deep learning. Sci Rep 7: 1–8, 2017.

    Article  Google Scholar 

  8. Liu S, Zheng H, Feng Y and Li W: Prostate cancer diagnosis using deep learning with 3D multiparametric MRI. Med Imaging 2017 Comput Diagnosis 10134: 1013428, 2017.

  9. Mehrtash A, Sedghi A, Ghafoorian M, Taghipour M, Tempany CM, Wells WM, Kapur T, Mousavi P, Abolmaesumi P and Fedorov A: Classification of clinical significance of MRI prostate findings using 3D convolutional neural networks. Med Imaging 2017 Comput Diagnosis 10134: 101342A, 2017.

  10. Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM and Thrun S: Dermatologist-level classification of skin cancer with deep neural networks. Nature 542: 115–118, 2017.

    Article  CAS  Google Scholar 

  11. Bae HJ, Kim CW, Kim N, Park BH, Kim N, Seo JB and Lee SM: A Perlin Noise-Based Augmentation Strategy for Deep Learning with Small Data Samples of HRCT Images. Sci Rep 8: 1–7, 2018.

    Google Scholar 

  12. Ding J, Chen B, Liu H and Huang M: Convolutional Neural Network with Data Augmentation for SAR Target Recognition. IEEE Geosci Remote Sens Lett 13: 364–368, 2016.

    Google Scholar 

  13. Lv JJ, Shao XH, Huang JS, Zhou XD and Zhou X: Data augmentation for face recognition. Neurocomputing 230: 184–196, 2017.

    Article  Google Scholar 

  14. Zhong Z, Zheng L, Kang G, Li S and Yang Y: Random Erasing Data Augmentation., 2017.

  15. Park SH, Goo JM and Jo CH: Receiver operating characteristic (ROC) curve: Practical review for radiologists. Korean J Radiol 5: 11–18, 2004.

    Article  Google Scholar 

  16. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D and Batra D: Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization. Int J Comput Vis 128: 336–359, 2020.

    Article  Google Scholar 

  17. Yoo S, Gujrathi I, Haider MA and Khalvati F: Prostate Cancer Detection using Deep Convolutional Neural Networks. Sci Rep 9: 1–10, 2019.

    Google Scholar 

  18. Glaister J, Cameron A, Wong A and Haider MA: Quantitative investigative analysis of tumour separability in the prostate gland using ultra-high b-value computed diffusion imaging. Proc Annu Int Conf IEEE Eng Med Biol Soc EMBS: 420–423, 2012.

  19. Khalvati F, Wong A and Haider MA: Automated prostate cancer detection via comprehensive multi-parametric magnetic resonance imaging texture feature models. BMC Med Imaging 15: 1–14, 2015.

    Article  Google Scholar 

  20. Khalvati F, Zhang J, Chung AG, Shafiee MJ, Wong A and Haider MA: MPCaD: A multi-scale radiomics-driven framework for automated prostate cancer localization and detection. BMC Med Imaging 18: 1–14, 2018.

    Article  Google Scholar 

  21. Nitish S, Geoffrey H, Alex K, Ilya S and Ruslan S: Dropout: A Simple Way to Prevent Neural Networks from Overfitting. J Mach Learn Res 15: 1929–1958, 2014.

    Google Scholar 

  22. Audhkhasi K, Osoba O and Kosko B: Noise-enhanced convolutional neural networks. Neural Networks 78: 15–23, 2016.

    Article  Google Scholar 

  23. Epstein JI, Egevad L, Amin MB, Delahunt B, Srigley JR and Humphrey PA: The 2014 international society of urological pathology (ISUP) consensus conference on gleason grading of prostatic carcinoma definition of grading patterns and proposal for a new grading system. Am J Surg Pathol 40: 244–252, 2016.

    Article  Google Scholar 

  24. Mottet N, Bellmunt J, Bolla M, … Cornford P: EAU-ESTRO-SIOG Guidelines on Prostate Cancer. Part 1: Screening, Diagnosis, and Local Treatment with Curative Intent. Eur Urol 71: 618–629, 2017.

  25. Parker JA, Kenyon R V. and Troxel DE: Comparison of Interpolating Methods for Image Resampling. IEEE Trans Med Imaging 2: 31–39, 1983.

    Article  Google Scholar 

  26. Namdar K, Gujrathi I, Haider MA and Khalvati F: Evolution-based Fine-tuning of CNNs for Prostate Cancer Detection. Int Conf Neural Inf Syst, 2019.

  27. Simonyan K and Zisserman A: Very deep convolutional networks for large-scale image recognition. 3rd Int Conf Learn Represent ICLR 2015 - Conf Track Proc: 1–14, 2015.

  28. Ruder S: An overview of gradient descent optimization algorithms. 1–14, 2016.

  29. Glorot X and Bengio Y: Understanding the difficulty of training deep feedforward neural networks. J Mach Learn Res 9: 249–256, 2010.

    Google Scholar 

  30. Anthon J. H: A Proof of the Conjecture That The Tukey-Kramer Multiple Comparisons Procedure Is Conservative. Ann Stat 12: 61–75, 1991.

    Google Scholar 

  31. Benjamini Y and Hochberg Y: Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc Ser B 57: 289–300, 1995.

    Google Scholar 

  32. McDonald JH: Handbook of Biolological Statistics. Sparky House Publishing, Baltimore, Maryland, U.S.A., 2014.

    Google Scholar 

  33. Vaishali S, Rao KK and Rao GVS: A review on noise reduction methods for brain MRI images. Int Conf Signal Process Commun Eng Syst - Proc SPACES 2015, Assoc with IEEE: 363–365, 2015.

  34. Islam MA, Jia S and Bruce NDB: How much Position Information Do Convolutional Neural Networks Encode? arXiv, 2020.

  35. Dyk DAV and Meng XL: The art of data augmentation. J Comput Graph Stat 10: 1–50, 2001.

    Article  Google Scholar 

Download references

Funding

This study received funding support in part by the Ontario Institute for Cancer Research, China Scholarship Council, and Chair in Medical Imaging and Artificial Intelligence, a joint Hospital-University Chair between the University of Toronto, The Hospital for Sick Children, and the SickKids Foundation.

Author information

Authors and Affiliations

  1. School of Optoelectronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan, China

    Ruqian Hao & Lin Liu

  2. Institute of Medical Science, University of Toronto, Toronto, ON, Canada

    Ruqian Hao & Farzad Khalvati

  3. Department of Diagnostic Imaging, The Hospital for Sick Children (SickKids), University of Toronto, 555 University Avenue, Toronto, ON, Canada

    Ruqian Hao, Khashayar Namdar & Farzad Khalvati

  4. Joint Department of Medical Imaging, Sinai Health System, University Health Network, University of Toronto, Toronto, ON, Canada

    Masoom A. Haider

  5. Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, ON, Canada

    Masoom A. Haider

  6. Sunnybrook Research Institute, Toronto, ON, Canada

    Masoom A. Haider

  7. Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON, Canada

    Farzad Khalvati

Authors
  1. Ruqian Hao
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Khashayar Namdar
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Lin Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Masoom A. Haider
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Farzad Khalvati
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Farzad Khalvati.

Ethics declarations

Conflict of Interest

The authors declare no competing interests.

Additional information

Publisher's Note

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

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hao, R., Namdar, K., Liu, L. et al. A Comprehensive Study of Data Augmentation Strategies for Prostate Cancer Detection in Diffusion-Weighted MRI Using Convolutional Neural Networks. J Digit Imaging 34, 862–876 (2021). https://doi.org/10.1007/s10278-021-00478-7

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10278-021-00478-7

Keywords

Search

Navigation