Advertisement

Building Disease Detection Algorithms with Very Small Numbers of Positive Samples

  • Ken C. L. Wong
  • Alexandros Karargyris
  • Tanveer Syeda-Mahmood
  • Mehdi MoradiEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10435)

Abstract

Although deep learning can provide promising results in medical image analysis, the lack of very large annotated datasets confines its full potential. Furthermore, limited positive samples also create unbalanced datasets which limit the true positive rates of trained models. As unbalanced datasets are mostly unavoidable, it is greatly beneficial if we can extract useful knowledge from negative samples to improve classification accuracy on limited positive samples. To this end, we propose a new strategy for building medical image analysis pipelines that target disease detection. We train a discriminative segmentation model only on normal images to provide a source of knowledge to be transferred to a disease detection classifier. We show that using the feature maps of a trained segmentation network, deviations from normal anatomy can be learned by a two-class classification network on an extremely unbalanced training dataset with as little as one positive for 17 negative samples. We demonstrate that even though the segmentation network is only trained on normal cardiac computed tomography images, the resulting feature maps can be used to detect pericardial effusion and cardiac septal defects with two-class convolutional classification networks.

References

  1. 1.
    Fei-Fei, L., Fergus, R., Perona, P.: One-shot learning of object categories. IEEE Trans. Pattern Anal. Mach. Intell. 28(4), 594–611 (2006)CrossRefGoogle Scholar
  2. 2.
    Geva, T., Martins, J.D., Wald, R.M.: Atrial septal defects. Lancet 383(9932), 1921–1932 (2014)CrossRefGoogle Scholar
  3. 3.
    Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., Bengio, Y.: Generative adversarial nets. In: Advances in Neural Information Processing Systems, pp. 2672–2680 (2014)Google Scholar
  4. 4.
    Kingma, D.P., Ba, J.L.: Adam: a method for stochastic optimization arXiv:1412.6980 [cs.LG] (2014)
  5. 5.
    Moradi, M., Gur, Y., Wang, H., Prasanna, P., Syeda-Mahmood, T.: A hybrid learning approach for semantic labeling of cardiac CT slices and recognition of body position. In: IEEE 13th International Symposium on Biomedical Imaging, pp. 1418–1421 (2016)Google Scholar
  6. 6.
    O’Leary, S.M., Williams, P.L., Williams, M.P., Edwards, A.J., Roobottom, C.A., Morgan-Hughes, G.J., Manghat, N.E.: Imaging the pericardium: appearances on ECG-gated 64-detector row cardiac computed tomography. Br. J. Radiol. 83(987), 194–205 (2010)CrossRefGoogle Scholar
  7. 7.
    Penny, D.J., Vick, G.W.: Ventricular septal defect. Lancet 377(9771), 1103–1112 (2011)CrossRefGoogle Scholar
  8. 8.
    Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). doi: 10.1007/978-3-319-24574-4_28 CrossRefGoogle Scholar
  9. 9.
    Shin, H.C., Roth, H.R., Gao, M., Lu, L., Xu, Z., Nogues, I., Yao, J., Mollura, D., Summers, R.M.: Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Trans. Med. Imaging 35(5), 1285–1298 (2016)CrossRefGoogle Scholar
  10. 10.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556 [cs.CV] (2014)
  11. 11.
    Socher, R., Ganjoo, M., Manning, C.D., Ng, A.: Zero-shot learning through cross-modal transfer. In: Advances in Neural Information Processing Systems, pp. 935–943 (2013)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Ken C. L. Wong
    • 1
  • Alexandros Karargyris
    • 1
  • Tanveer Syeda-Mahmood
    • 1
  • Mehdi Moradi
    • 1
    Email author
  1. 1.IBM Research – Almaden Research CenterSan JoseUSA

Personalised recommendations