Abstract
Nowadays, the huge volume of medical images represents an enormous challenge towards health-care organizations, as it is often hard for clinicians and researchers to manage, access, and share the image database easily. Content-based medical image retrieval (CBMIR) techniques are employed to facilitate the above process. It is known that a few concrete factors, including visual attributes extracted from images, measures encoding the similarity between images, user interaction, etc. play important roles in determining the retrieval performance. This paper concentrates on the similarity learning problem of CBMIR. A novel similarity learning paradigm is proposed via relative comparison, and a large database composed of 5,000 images is utilized to evaluate the retrieval performance. Extensive experimental results and comprehensive statistical analysis demonstrate the superiority of adopting the newly introduced learning paradigm, compared with several conventional supervised and semi-supervised similarity learning methods, in the presented CBMIR application.
Similar content being viewed by others
References
Frost, Sullivan: US Data storage management markets for healthcare. Online resource: http://www.frost.com
ARRS GoldMiner: Online resource: http://goldminer.arrs.org/home.php
Hearst M, Divoli A, Guturu H, Ksikes A, Nakov P, Wooldridge M, Ye J: BioText search engine: beyond abstract search. Bioinformatics 23(16):2196–2197, 2007
iMedline: Online resource: http://archive.nlm.nih.gov/iti/
Müller H, Michoux N, Bandon D, Geissbuhler A: A review of content-based image retrieval systems in medical applications - clinical benefits and future directions. Int J Med Inform 73(1):1–23, 2004
Müller H, Deselaers T: Tutorial on medical image retrieval: content-based image retrieval. Med Inf Europe 2005. Online resource: http://thomas.deselaers.de/teaching/files/tutorial_mie06/ContentBasedImageRetrieval.pdf
Tagare H, Jaffe C, Duncan J: Medical image databases: a content-based retrieval approach. J Am Med Inform Assoc 4(3):184–198, 1997
Ghosh P, Antani S, Long R, Thoma G: Review of medical image retrieval systems and future directions. IEEE Int Symp Computer-Based Med Syst 1:1–16, 2011
Sweet H, Mutalik P, Neklesa V, Horvath L, Lee C, Richter J, Tocino I, Fisher P: Voice-activated retrieval of mammography reference images. J Digit Imaging 11(2):65–73, 1998
Oh J, Yang Y, El Naqa I: Adaptive learning for relevance feedback: application to digital mammography. Med Phys 37(8):4432–4444, 2010
El Naqa I, Yang Y, Galatsanos N, Nishikawa R, Wernick M: A similarity learning approach to content-based image retrieval: application to digital mammography. IEEE Trans Med Imaging 23(10):1233–1244, 2004
Kinoshita S, de Azevedo-Marques P, Pereira R, Rodrigues J, Rangayyan R: Content-based retrieval of mammograms using visual features related to breast density patterns. J Digit Imaging 20(2):172–190, 2007
Tourassi G, Harrawood B, Singh S, Lo J, Floyd C: Evaluation of information-theoretic similarity measures for content-based retrieval and detection of masses in mammography. Med Phys 34(1):140–150, 2007
Zheng B, Lu A, Hardesty L, Sumkin J, Hakim C, Ganott M, Gur D: A Method to improve visual similarity of breast masses for an interactive computer-aided diagnosis environment. Med Phys 33(1):111–117, 2006
Mazurowski M, Habas P, Zurada J, Tourassi G: Decision optimization of case-based computer-aided decision systems using genetic algorithms with application to mammography. Phys Med Biol 53(4):895–908, 2008
Dy J, Brodley C, Kak A, Broderick L, Aisen A: Unsupervised feature selection applied to content-based retrieval of lung images. IEEE Trans on Pattern Anal Mach Intell 25(3):373–378, 2003
Shyu C, Brodley C, Kak A, Kosaka A, Aisen A, Broderick L: ASSERT: a physician-in-the-loop content-based retrieval system for HRCT image databases. Comp Vision Image Underst 75(1–2):111–132, 1999
Lehmann T, Güld M, Thies C, Fischer B, Spitzer K, Keysers D, Ney H, Kohnen M, Schubert H, Wein B: Content-based image retrieval in medical applications. Methods Inf Med 43:354–361, 2004
Huang W, Chan K, Li H, Lim J, Liu J, Wong T: A computer assisted method for nuclear cataract grading from slit-lamp images using ranking. IEEE Trans Med Imaging 30(1):94–107, 2011
Yang L, Jin R: Distance metric learning: a comprehensice survey. Online resource: http://www.cs.cmu.edu/liuy/frame_survey_v2.pdf
Guo G, Jain A, Ma W, Zhang H: Learning similarity measure for natural image retrieval with relevance feedback. IEEE Trans Neural Netw 13:811–820, 2002
Hertz T, Bar-Hillel A, Weinshall D: Learning distance functions for image retrieval. IEEE Conf Comput Vis Pattern Recognit 2:570–577, 2004
Si L, Jin R, Hoi S, Lyu M: Collaborative image retrieval via regularized metric learning. Mulitmedia Systems 12:34–44, 2006
Globerson A, Roweis S: Metric learning by collapsing classes. Adv Neural Inf Process Syst 1:451–458, 2005
Weinberger K, Blitzer J, Saul L: Distance metric learning for large margin nearest neighbor classification. Adv Neural Inf Process Syst 1:1473–1480, 2006
Zhu X: Semi-supervised learning literature survey, Online resource: citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.99.9681&rep=rep1&type=pdf
Chapelle O, Scholkopf B, Zien A: Semi-supervised learning. MIT Press, 2006
Xing E, Ng A, Jordan M, Russell S: Distance metric learning, with application to clustering with side-information. Adv Neural Inf Process Syst 1:505–512, 2003
Schultz M, Joachims T: Learning a distance metric from relative comparisons. Adv Neural Inf Process Syst 1:41–48, 2004
Schroff F: Harvesting image databases from the web. IEEE Trans Pattern Anal Mach Intell 33(4):754–766, 2011
Jing Y: VisualRank: applying pagerank to large-scale image search. IEEE Trans Pattern Anal Mach Intell 30(11):1877–1890, 2008
Siddiquie B: Image ranking and retrieval based on multi-attribute queries. IEEE Conf Comput Vis Pattern Recognit 1:801–808, 2011
Datta R, Joshi D, Li J, Wang J: Image retrieval: ideas, influences, and trends of the new age. ACM Comput Surv 40(2):5:1–5:60, 2008
Kendall M: A new measure of rank correlation. Biometrika 30:81–93, 1938
World Health Organization: The world health report: life in the 21st century - a vision for all. Online resource: http://www.who.int/whr/1998/en/whr98_en.pdf, 1998
Klein B, Klein R, Linton K, Magli Y, Neider M: Assessment of cataracts from photographs in the Beaver dam eye study. Ophthalmology 97(11):1428–1433, 1990
Li C, Kao C, Gore J, Ding Z: Implicit active contours driven by local binary fitting energy. IEEE Conf Comput Vis Pattern Recognit 1:1–7, 2007
Rice J: Mathematical Statistics and Data Analysis, 2nd edition. Duxbury Press, 2007
Gold C, Sollich P: Model selection for support vector machine classification. Neurocomputing 55:221–249, 2003
Weston J: Leave-one-out support vector machines. Int Joint Conf Artif Intell 1:727–733, 1999
Chapelle O, Vapnik V, Bousquet O, Mukherjee S: Choosing multiple parameters for support vector machines. Mach Learn 46(1–3):131–159, 2002
Weinberger K: Large margin NN classifier - matlab toolbox. Online resource http://www.weinbergerweb.net/Downloads/LMNN.html
Chapelle O: Selection of kernel parameters - matlab toolbox. Online resource http://olivier.chapelle.cc/ams/
Congdon N, Chang N, Botelho P, Stark W, Datiles III M: Cataract: clinical types. Duane’s Ophthalmology, 2006 edition, vol 1, chapter 73(B). Lippincott: Williams & Wilkins, 2006
Zigler Jr J, Datiles III M: Pathogenesis of cataracts. Duane’s Ophthalmology, 2006 edition, vol 1, chapter 72(B). Lippincott: Williams & Wilkins, 2006
Author information
Authors and Affiliations
Corresponding authors
Appendix
Appendix
In Eq. 6, the second term \(\frac {\text {exp}\big (2(\ell _{(q,y)} - \ell _{(q,x)})\big )-1}{\text {exp}\big (2(\ell _{(q,y)} - \ell _{(q,x)})\big )+1}\) of the right-hand side (RHS) of SKT can be viewed as a coefficient in the derivation, since it is not related to parameters a to learn in this study. Thus, the following derivation only focuses on the first term \(\frac {\text {exp}\big (2(s_{(q,x)} - s_{(q,y)})\big )-1}{\text {exp}\big (2(s_{(q,x)} - s_{(q,y)})\big )+1}\) of RHS in SKT. For the ease of writing, let us denote the second term \(\frac {\text {exp}\big (2(\ell _{(q,y)} - \ell _{(q,x)})\big )-1}{\text {exp}\big (2(\ell _{(q,y)} - \ell _{(q,x)})\big )+1}\) as term coeff. After differentiation, Eq. 6 becomes:
Then, after replacing the term coeff with its original mathematical form, the gradient can be rewritten as:
which is the same as Eq. 7.
Rights and permissions
About this article
Cite this article
Huang, W., Zhang, P. & Wan, M. A Novel Similarity Learning Method via Relative Comparison for Content-Based Medical Image Retrieval. J Digit Imaging 26, 850–865 (2013). https://doi.org/10.1007/s10278-013-9591-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10278-013-9591-x