Learning from Only Positive and Unlabeled Data to Detect Lesions in Vascular CT Images

  • Maria A. Zuluaga
  • Don Hush
  • Edgar J. F. Delgado Leyton
  • Marcela Hernández Hoyos
  • Maciej Orkisz
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6893)


Detecting vascular lesions is an important task in the diagnosis and follow-up of the coronary heart disease. While most existing solutions tackle calcified and non-calcified plaques separately, we present a new algorithm capable of detecting both types of lesions in CT images. It builds up on a semi-supervised classification framework, in which the training set is made of both unlabeled data and a small amount of data labeled as normal. Our method takes advantage of the arrival of newly acquired data to re-train the classifier and improve its performance. We present results on synthetic data and on datasets from 15 patients. With a small amount of labeled training data our method achieved a 89.8% true positive rate, which is comparable to state-of-the-art supervised methods, and the performance can improve after additional iterations.


Support Vector Machine Synthetic Data True Positive Rate Unlabeled Data Empirical Risk 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Pugliese, F., Hunink, M.G.M., Gruszczynska, K., Alberghina, F., Malagó, R., van Pelt, N., Mollet, N.R., Cademartiri, F., Weustink, A.C., Meijboom, W.B., Witteman, C.L.M., de Feyter, P.J., Krestin, G.P.: Learning curve for coronary CT angiography: what constitutes sufficient training? Radiology 251, 359–368 (2009)CrossRefGoogle Scholar
  2. 2.
    Saur, S.C., Alkadhi, H., Desbiolles, L., Székely, G., Cattin, P.C.: Automatic detection of calcified coronary plaques in computed tomography data sets. In: Metaxas, D., Axel, L., Fichtinger, G., Székely, G. (eds.) MICCAI 2008, Part I. LNCS, vol. 5241, pp. 170–177. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  3. 3.
    Makrogiannis, S., Bhotika, R., Miller, J.V., Skinner Jr., J., Vass, M.: Nonparametric intensity priors for level set segmentation of low contrast structures. In: Yang, G.-Z., Hawkes, D., Rueckert, D., Noble, A., Taylor, C. (eds.) MICCAI 2009, Part I. LNCS, vol. 5761, pp. 239–246. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  4. 4.
    Tessmann, M., Vega-Higuera, F., Fritz, D., Scheuering, M., Greiner, G.: Multi-scale feature extraction for learning-based classification of coronary artery stenosis. In: Proc. SPIE Med. Imaging, vol. 7260 (2009), doi:10.1117/12.811639Google Scholar
  5. 5.
    Zuluaga, M.A., Magnin, I.E., Hernández Hoyos, M., Delgado Leyton, E.J.F., Lozano, F., Orkisz, M.: Automatic detection of abnormal vascular cross-sections based on Density Level Detection and Support Vector Machines. Int. J. Computer Assisted Radiol. Surg. 6, 163–174 (2011)CrossRefGoogle Scholar
  6. 6.
    Porter, R., Ruggiero, C., Hush, D.: Density-based similarity measures for content based search. In: 43rd Asilomar Conf. Signals Systems Computers, pp. 390–394 (2009)Google Scholar
  7. 7.
    Steinwart, I., Hush, D., Scovel, C.: A classification framework for anomaly detection. J. Mach. Learning Research 6, 211–232 (2005)MathSciNetzbMATHGoogle Scholar
  8. 8.
    Zuluaga, M.A., Delgado Leyton, E.J.F., Hernández Hoyos, M., Orkisz, M.: Feature selection for SVM-based vascular anomaly detection. In: Menze, B., Langs, G., Tu, Z., Criminisi, A. (eds.) Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging. LNCS, vol. 6533, pp. 141–152. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  9. 9.
    Lesage, D., Angelini, E.D., Bloch, I., Funka-Lea, G.: A review of 3D vessel lumen segmentation techniques: models, features and extraction schemes. MedIA 13(6), 819–845 (2009)Google Scholar
  10. 10.
    Chang, C.-C., Lin, C.-J.: LIBSVM: a library for support vector machines (2004),
  11. 11.
    Schaap, M., Metz, C., van Walsum, T., van der Giessen, A.G., Weustink, A.C., Mollet, N.R., Bauer, C., Bogunovic, H., Castro, C., Deng, X., Dikici, E., O’Donnell, T., Frenay, M., Friman, O., Hernández Hoyos, M., Kitslaar, P.H., Krissian, K., Kühnel, C., Luengo-Oroz, M.A., Orkisz, M., Smedby, Ö., Styner, M., Szymczak, A., Tek, H., Wang, C., Warfield, S.K., Zambal, S., Zhang, Y., Krestin, G.P., Niessen, W.J.: Standardized evaluation methodology and reference database for evaluating coronary artery centerline extraction algorithms. MedIA 13(5), 704–715 (2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Maria A. Zuluaga
    • 1
    • 2
  • Don Hush
    • 3
  • Edgar J. F. Delgado Leyton
    • 1
    • 2
  • Marcela Hernández Hoyos
    • 2
  • Maciej Orkisz
    • 1
  1. 1.INSERM U1044CREATIS; Université de Lyon; Université Lyon 1; INSA-Lyon; CNRS UMR5220VilleurbanneFrance
  2. 2.Grupo Imagine, Grupo de Ingeniería BiomédicaUniversidad de los AndesBogotáColombia
  3. 3.Los Alamos National LaboratoryLos AlamosUSA

Personalised recommendations