Fast Prospective Detection of Contrast Inflow in X-ray Angiograms with Convolutional Neural Network and Recurrent Neural Network

  • Hua MaEmail author
  • Pierre Ambrosini
  • Theo van Walsum
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10435)


Automatic detection of contrast inflow in X-ray angiographic sequences can facilitate image guidance in computer-assisted cardiac interventions. In this paper, we propose two different approaches for prospective contrast inflow detection. The methods were developed and evaluated to detect contrast frames from X-ray sequences. The first approach trains a convolutional neural network (CNN) to distinguish whether a frame has contrast agent or not. The second method extracts contrast features from images with enhanced vessel structures; the contrast frames are then detected based on changes in the feature curve using long short-term memory (LSTM), a recurrent neural network architecture. Our experiments show that both approaches achieve good performance on detection of the beginning contrast frame from X-ray sequences and are more robust than a state-of-the-art method. As the proposed methods work in prospective settings and run fast, they have the potential of being used in clinical practice.



This work was supported by Technology Foundation STW, IMAGIC project under the iMIT program (grant number 12703).

Conflict of interest. The authors declare that they have no conflict of interest.


  1. 1.
    Ruijters, D., ter Haar Romeny, B.M., Suetens, P.: Vesselness-based 2D–3D registration of the coronary arteries. Int. J. Comput. Assist. Radiol. Surg. 4(4), 391–397 (2009)CrossRefGoogle Scholar
  2. 2.
    Baka, N., Lelieveldt, B.P.F., Schultz, C., Niessen, W., van Walsum, T.: Respiratory motion estimation in X-ray angiography for improved guidance during coronary interventions. Phys. Med. Biol. 60(9), 3617 (2015)CrossRefGoogle Scholar
  3. 3.
    Condurache, A., Aach, T., Eck, K., Bredno, J.: Fast detection and processing of arbitrary contrast agent injections in coronary angiography and fluoroscopy. In: Tolxdorff, T., Braun, J., Handels, H., Horsch, A., Meinzer, H.P. (eds.) Bildverarbeitung füer die Medizin, pp. 5–9. Springer, Heidelberg (2004). doi: 10.1007/978-3-642-18536-6_2 CrossRefGoogle Scholar
  4. 4.
    Liao, R., You, W., Liu, Y., Yan, M., John, M., Shea, S.: Integrated spatiotemporal analysis for automatic contrast agent inflow detection on angiography and fluoroscopy during transcatheter aortic valve implantation. Med. Phys. 40(4), 041914-1–041914-11 (2013)CrossRefGoogle Scholar
  5. 5.
    Liao, R., Miao, S., Zheng, Y.: Automatic and efficient contrast-based 2-D/3-D fusion for trans-catheter aortic valve implantation (TAVI). Comput. Med. Imaging Graph. 37(2), 150–161 (2013)CrossRefGoogle Scholar
  6. 6.
    You, W., Liao, R., Yan, M., John, M.: Spatio-temporal analysis for automatic contrast injection detection on angiography during trans-catheter aortic valve implantation. In: ISBI, pp. 702–706 (2011)Google Scholar
  7. 7.
    Zhao, X., Miao, S., Du, L., Liao, R.: Robust 2-D/3-D registration of CT volumes with contrast-enhanced X-ray sequences in electrophysiology based on a weighted similarity measure and sequential subspace optimization. In: IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 934–938 (2013)Google Scholar
  8. 8.
    Chen, T., Funka-Lea, G., Comaniciu, D.: Robust and fast contrast inflow detection for 2D X-ray fluoroscopy. In: Fichtinger, G., Martel, A., Peters, T. (eds.) MICCAI 2011. LNCS, vol. 6891, pp. 243–250. Springer, Heidelberg (2011). doi: 10.1007/978-3-642-23623-5_31 CrossRefGoogle Scholar
  9. 9.
    Hoffmann, M., Müller, S., Kurzidim, K., Strobel, N., Hornegger, J.: Robust identification of contrasted frames in fluoroscopic images. In: Handels, H., Deserno, T., Meinzer, H.P., Tolxdorff, T. (eds.) Bildverarbeitung für die Medizin 2015, pp. 23–28. Springer, Heidelberg (2015)CrossRefGoogle Scholar
  10. 10.
    LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015)CrossRefGoogle Scholar
  11. 11.
    Greenspan, H., van Ginneken, B., Summers, R.M.: Guest editorial deep learning in medical imaging: overview and future promise of an exciting new technique. IEEE Trans. Med. Imaging 35(5), 1153–1159 (2016)CrossRefGoogle Scholar
  12. 12.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  13. 13.
    Ma, H., Hoogendoorn, A., Regar, E., Niessen, W.J., van Walsum, T.: Automatic online layer separation for vessel enhancement in X-ray angiograms for percutaneous coronary interventions. Med. Image Anal. 39, 145–161 (2017)CrossRefGoogle Scholar
  14. 14.
    Frangi, A.F., Niessen, W.J., Vincken, K.L., Viergever, M.A.: Multiscale vessel enhancement filtering. In: Wells, W.M., Colchester, A., Delp, S. (eds.) MICCAI 1998. LNCS, vol. 1496, pp. 130–137. Springer, Heidelberg (1998). doi: 10.1007/BFb0056195 CrossRefGoogle Scholar
  15. 15.
    Feng, J., Xu, H., Yan, S.: Online robust PCA via stochastic optimization. In: NIPS, pp. 404–412 (2013)Google Scholar
  16. 16.
    Ma, H., Dibildox, G., Banerjee, J., Niessen, W., Schultz, C., Regar, E., Walsum, T.: Layer separation for vessel enhancement in interventional X-ray angiograms using morphological filtering and robust PCA. In: Linte, C.A., Yaniv, Z., Fallavollita, P. (eds.) AE-CAI 2015. LNCS, vol. 9365, pp. 104–113. Springer, Cham (2015). doi: 10.1007/978-3-319-24601-7_11 CrossRefGoogle Scholar
  17. 17.
    Rodriguez, P., Wohlberg, B.: Fast principal component pursuit via alternating minimization. In: Image Processing, IEEE International Conference on, 69–73, (2013)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Biomedical Imaging Group RotterdamErasmus MCRotterdamThe Netherlands

Personalised recommendations