Abstract
We address the problem of instance label stability in multiple instance learning (MIL) classifiers. These classifiers are trained only on globally annotated images (bags), but often can provide fine-grained annotations for image pixels or patches (instances). This is interesting for computer aided diagnosis (CAD) and other medical image analysis tasks for which only a coarse labeling is provided. Unfortunately, the instance labels may be unstable. This means that a slight change in training data could potentially lead to abnormalities being detected in different parts of the image, which is undesirable from a CAD point of view. Despite MIL gaining popularity in the CAD literature, this issue has not yet been addressed. We investigate the stability of instance labels provided by several MIL classifiers on 5 different datasets, of which 3 are medical image datasets (breast histopathology, diabetic retinopathy and computed tomography lung images). We propose an unsupervised measure to evaluate instance stability, and demonstrate that a performance-stability trade-off can be made when comparing MIL classifiers.
Chapter PDF
References
Andrews, S., Tsochantaridis, I., Hofmann, T.: Support vector machines for multiple-instance learning. In: NIPS, pp. 561–568 (2002)
Ben-Hur, A., Elisseeff, A., Guyon, I.: A stability based method for discovering structure in clustered data. In: Pac. Symp. Biocomput., pp. 6–17 (2001)
Bi, J., Liang, J.: Multiple instance learning of pulmonary embolism detection with geodesic distance along vascular structure. In: CVPR, pp. 1–8 (2007)
Chen, Y., Bi, J., Wang, J.: MILES: Multiple-instance learning via embedded instance selection. IEEE T. Pattern. Anal. Mach. Intel. 28(12), 1931–1947 (2006)
Cheplygina, V., Tax, D.M.J., Loog, M.: Multiple instance learning with bag dissimilarities. Pattern Recognition 48(1), 264–275 (2015)
Cheplygina, V., et al.: Classification of COPD with multiple instance learning. In: ICPR, pp. 1508–1513 (2014)
Decencière, E., et al.: Feedback on a publicly distributed image database: the Messidor database. Image Anal. Stereol., 231–234 (2014)
Dundar, M.M., Fung, G., et al.: Multiple-instance learning algorithms for computer-aided detection. IEEE T. Biomed. Eng. 55(3), 1015–1021 (2008)
Kandemir, M., Hamprecht, F.A.: Computer-aided diagnosis from weak supervision: A benchmarking study. Comput. Med. Imag. Grap. (2014) (in press)
Kandemir, M., Zhang, C., Hamprecht, F.A.: Empowering multiple instance histopathology cancer diagnosis by cell graphs. In: Golland, P., Hata, N., Barillot, C., Hornegger, J., Howe, R. (eds.) MICCAI 2014, Part II. LNCS, vol. 8674, pp. 228–235. Springer, Heidelberg (2014)
Liang, J., Bi, J.: Computer aided detection of pulmonary embolism with tobogganing and mutiple instance classification in CT pulmonary angiography. In: Karssemeijer, N., Lelieveldt, B. (eds.) IPMI 2007. LNCS, vol. 4584, pp. 630–641. Springer, Heidelberg (2007)
Marques, J.: Osteoarthritis imaging by quantification of tibial trabecular bone. Ph.D. thesis, Københavns Universitet (2013)
Melendez, J., et al.: A novel multiple-instance learning-based approach to computer-aided detection of tuberculosis on chest x-rays. TMI 31(1), 179–192 (2014)
Pedersen, J.H., et al.: The Danish randomized lung cancer CT screening trial-overall design and results of the prevalence round. J. Thorac. Oncol. 4(5), 608–614 (2009)
Poggio, T., Rifkin, R., Mukherjee, S., Niyogi, P.: General conditions for predictivity in learning theory. Nature 428(6981), 419–422 (2004)
Quellec, G., et al.: A multiple-instance learning framework for diabetic retinopathy screening. MedIA 16(6), 1228–1240 (2012)
Sørensen, L., Nielsen, M., Lo, P., Ashraf, H., Pedersen, J.H., de Bruijne, M.: Texture-based analysis of COPD: a data-driven approach. TMI 31(1), 70–78 (2012)
Sun, L., Lu, Y., Yang, K., Li, S.: ECG analysis using multiple instance learning for myocardial infarction detection. IEEE T. Biomed. Eng. 59(12), 3348–3356 (2012)
Viola, P., Platt, J., Zhang, C.: Multiple instance boosting for object detection. In: NIPS, pp. 1417–1424 (2005)
Wang, S., et al.: Seeing is believing: Video classification for computed tomographic colonography using multiple-instance learning. TMI 31(5), 1141–1153 (2012)
Wu, D., Bi, J., Boyer, K.: A min-max framework of cascaded classifier with multiple instance learning for computer aided diagnosis. In: CVPR, pp. 1359–1366 (2009)
Xu, Y., et al.: Weakly supervised histopathology cancer image segmentation and classification. MedIA 18(3), 591–604 (2014)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Cheplygina, V., Sørensen, L., Tax, D.M.J., de Bruijne, M., Loog, M. (2015). Label Stability in Multiple Instance Learning. In: Navab, N., Hornegger, J., Wells, W., Frangi, A. (eds) Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015. MICCAI 2015. Lecture Notes in Computer Science(), vol 9349. Springer, Cham. https://doi.org/10.1007/978-3-319-24553-9_66
Download citation
DOI: https://doi.org/10.1007/978-3-319-24553-9_66
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-24552-2
Online ISBN: 978-3-319-24553-9
eBook Packages: Computer ScienceComputer Science (R0)