A Comparative Assessment of Feed-Forward and Convolutional Neural Networks for the Classification of Prostate Lesions

  • Sabrina Marnell
  • Patrick RileyEmail author
  • Ivan Olier
  • Marc Rea
  • Sandra Ortega-Martorell
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11872)


Prostate cancer is the most common cancer in men in the UK. An accurate diagnosis at the earliest stage possible is critical in its treatment. Multi-parametric Magnetic Resonance Imaging is gaining popularity in prostate cancer diagnosis, it can be used to actively monitor low-risk patients, and it is convenient due to its non-invasive nature. However, it requires specialist knowledge to review the abundance of available data, which has motivated the use of machine learning techniques to speed up the analysis of these many and complex images. This paper focuses on assessing the capabilities of two neural network approaches to accurately discriminate between three tissue types: significant prostate cancer lesions, non-significant lesions, and healthy tissue. For this, we used data from a previous SPIE ProstateX challenge that included significant and non-significant lesions, and we extended the dataset to include healthy prostate tissue due to clinical interest. Feed-Forward and Convolutional Neural Networks have been used, and their performances were evaluated using 80/20 training/test splits. Several combinations of the data were tested under different conditions and summarised results are presented. Using all available imaging data, a Convolutional Neural Network three-class classifier comparing prostate lesions and healthy tissue attains an Area Under the Curve of 0.892.


Feed-forward neural networks Convolutional neural networks SPIE ProstateX mpMRI Prostate cancer 



This work has been funded by the LJMU Scholarship Fund.


  1. 1.
    Prostate cancer screening scan hope - BBC NewsGoogle Scholar
  2. 2.
    An, J.Y., Sidana, A., Choyke, P.L., Wood, B.J., Pinto, P.A., Türkbey, İ.B.: Multiparametric magnetic resonance imaging for active surveillance of prostate cancer. Balkan Med. J. 34, 388–396 (2017). Scholar
  3. 3.
    McCulloch, W.S., Pitts, W.: A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophys. 5, 115–133 (1943). Scholar
  4. 4.
    Srivastava, N., Hinton, G., Krizhevsky, A., Salakhutdinov, R.: Dropout: A Simple Way to Prevent Neural Networks from Overfitting (2014)Google Scholar
  5. 5.
    Riley, P., Olier, I., Rea, M., Lisboa, P., Ortega-Martorell, S.: A voting ensemble method to assist the diagnosis of prostate cancer using multiparametric MRI. In: Vellido, A., Gibert, K., Angulo, C., Martín Guerrero, J.D. (eds.) WSOM 2019. AISC, vol. 976, pp. 294–303. Springer, Cham (2020). Scholar
  6. 6.
    Litjens, G., Debats, O., Barentsz, J., Karssemeijer, N., Huisman, H.: Computer-aided detection of prostate cancer in MRI. IEEE Trans. Med. Imaging 33, 1083–1092 (2014). Scholar
  7. 7.
    Gallagher, J.: Prostate cancer treatment “not always needed” - BBC News/ Health (2016).
  8. 8.
    Chen, Q., Xu, X., Hu, S., Li, X., Zou, Q., Li, Y.: A transfer learning approach for classification of clinical significant prostate cancers from mpMRI scans. In: Armato III, S.G., Petrick, N.A. (eds.) SPIE Medical Imaging 2017: Computer-Aided Diagnosis, p. 101344F. International Society for Optics and Photonics, Orlando (2017)Google Scholar
  9. 9.
    Kitchen, A., Seah, J.: Support vector machines for prostate lesion classification. In: Armato III, S.G., Petrick, N.A. (eds.) SPIE Medical Imaging 2017: Computer-Aided Diagnosis, p. 1013427. International Society for Optics and Photonics, Orlando (2017)Google Scholar
  10. 10.
    Seah, J.C.Y., Tang, J.S.N., Kitchen, A.: Detection of prostate cancer on multiparametric MRI. In: Armato, S.G., Petrick, N.A. (eds.) SPIE Medical Imaging 2017: Computer-Aided Diagnosis. p. 1013429. International Society for Optics and Photonics (2017)Google Scholar
  11. 11.
    Langer, D.L., et al.: Prostate tissue composition and MR measurements: investigating the relationships between ADC, T2, Ktrans, ve, and corresponding histologic features. Radiology 255, 485–494 (2010). Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Applied MathematicsLiverpool John Moores UniversityLiverpoolUK
  2. 2.Clatterbridge Cancer Centre NHS Foundation TrustWirralUK

Personalised recommendations