All Title Author
Keywords Abstract

Restricted Boltzmann Machines for Classification of Hepatocellular Carcinoma

DOI: 10.1155/2014/418069

Full-Text   Cite this paper   Add to My Lib


Multiple antigen miniarrays can provide accurate tools for cancer detection and diagnosis. These miniarrays can be validated by examining their operating characteristics in classifying individuals as either cancer patients or normal (non-cancer) subjects. We describe the use of restricted Boltzmann machines for this classification problem, relative to diagnosis of hepatocellular carcinoma. In this setting, we find that its operating characteristics are similar to a logistic regression standard and suggest that restricted Boltzmann machines merit further consideration for classification problems. 1. Introduction We have previously investigated the utility of antibody profiles to seven tumor-associated antigens (TAAs) for discriminating between cancer patients and controls [1]. We found that these multiple antigen miniarrays could provide accurate tools for cancer detection and diagnosis and suggested that performance of the miniarrays might be enhanced by other combinations of TAAs appropriately selected for different cancer cohorts. We return to this theme in the present paper, where we examine the utility of an expanded panel of 12 antibody profiles for cancer diagnosis of hepatocellular carcinoma (HCC), based on serum samples from newly diagnosed HCC patients and normal controls. To this end, we apply a fairly recent approach, restricted Boltzmann machines [2–5], to the classification problem at hand. For comparative purposes, we also utilize logistic regression, to provide a baseline of discriminative performance. Our aim here is to determine whether restricted Boltzmann machines provide a viable technique for classification of HCC, compared to the logistic regression standard. 2. Materials and Methods 2.1. Sera Samples In all, sera samples from 175 HCC patients and 90 normal controls were amassed, as follows. Sera from 76 patients with HCC from Xiamen in China were obtained from the serum bank of the Cancer Autoimmunity Research Laboratory at the University of Texas (El Paso, Texas, USA), which were originally provided by a collaborator in Sun Yat-sen University (Guangzhou, China). 84 HCC patients’ sera were collected from Korea and 15 from Japan. Ninety normal human sera (NHS) were originally obtained from the serum bank of the Autoimmune Disease Center at the Scripps Research Institute (La Jolla, CA, USA). All cancer patients were diagnosed according to established criteria; their serum samples were collected at the time of initial cancer diagnosis, when the patients had not received treatment with any chemotherapy or radiation therapy. Normal


[1]  J. A. Koziol, J.-Y. Zhang, C. A. Casiano et al., “Recursive partitioning as an approach to selection of immune markers for tumor diagnosis,” Clinical Cancer Research, vol. 9, no. 14, pp. 5120–5126, 2003.
[2]  G. E. Hinton, “Training products of experts by minimizing contrastive divergence,” Neural Computation, vol. 14, no. 8, pp. 1771–1800, 2002.
[3]  G. E. Hinton and R. R. Salakhutdinov, “Reducing the dimensionality of data with neural networks,” Science, vol. 313, no. 5786, pp. 504–507, 2006.
[4]  A. Fischer and C. Igel, “An introduction to restricted Boltzmann machines,” in Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, L. Alvarez, Ed., vol. 7441 of Lecture Notes in Computer Science, pp. 14–36, Springer, Berlin, Germany, 2012.
[5]  A. Fischer and C. Igel, “Training restricted Boltzmann machines: an introduction,” Pattern Recognition, vol. 47, pp. 25–39, 2014.
[6]  J.-Y. Zhang, E. K. L. Chan, X.-X. Peng, and E. M. Tan, “A novel cytoplasmic protein with RNA-binding motifs is an autoantigen in human hepatocellular carcinoma,” The Journal of Experimental Medicine, vol. 189, no. 7, pp. 1101–1110, 1999.
[7]  F. Müeller-Pillasch, U. Lacher, C. Wallrapp et al., “Cloning of a gene highly overexpressed in cancer coding for a novel KH-domain containing protein,” Oncogene, vol. 14, no. 22, pp. 2729–2733, 1997.
[8]  J. Nielsen, J. Christiansen, J. Lykke-Andersen, A. H. Johnsen, U. M. Wewer, and F. C. Nielsen, “A family of insulin-like growth factor II mRNA-binding proteins represses translation in late development,” Molecular and Cellular Biology, vol. 19, no. 2, pp. 1262–1270, 1999.
[9]  H. Imai, E. K. L. Chan, K. Kiyosawa, X.-D. Fu, and E. M. Tan, “Novel nuclear autoantigen with splicing factor motifs identified with antibody from hepatocellular carcinoma,” The Journal of Clinical Investigation, vol. 92, no. 5, pp. 2419–2426, 1993.
[10]  K. Looi, R. Megliorino, F.-D. Shi, X.-X. Peng, Y. Chen, and J.-Y. Zhang, “Humoral immune response to p16, a cyclin-dependent kinase inhibitor in human malignancies,” Oncology Reports, vol. 16, no. 5, pp. 1105–1110, 2006.
[11]  L. S. Hoo, J. Y. Zhang, and E. K. L. Chan, “Cloning and characterization of a novel 90 kDa “companion” auto-antigen of p62 overexpressed in cancer,” Oncogene, vol. 21, no. 32, pp. 5006–5015, 2002.
[12]  H. Larochelle and Y. Bengio, “Classification using discriminative restricted boltzmann machines,” in Proceedings of the 25th International Conference on Machine Learning, pp. 536–543, ACM, New York, NY, USA, July 2008.
[13]  A. Y. Ng and M. I. Jordan, “On discriminative vs. generative classifiers: a comparison of logistic regression and na?ve Bayes,” in Advances in Neural Information Processing Systems, T. G. Diettrich, S. Becker, and Z. Ghahramani, Eds., pp. 841–848, MIT Press, Cambridge, Mass, USA, 2001.
[14]  G. E. Hinton, “A practical guide to training restricted Boltzmann machines,” Tech. Rep. UTML TR 2010-003, Department of Computer Science, University of Toront, 2010,
[15]  E. M. Tan, “Autoantibodies as reporters identifying aberrant cellular mechanisms in tumorigenesis,” The Journal of Clinical Investigation, vol. 108, no. 10, pp. 1411–1415, 2001.
[16]  E. M. Tan and J. Y. Zhang, “Autoantibodies to tumor-associated antigens: reporters from the immune system,” Immunological Reviews, vol. 222, no. 1, pp. 328–340, 2008.
[17]  O. Naggara, J. Raymond, F. Guilbert, D. Roy, A. Weill, and D. G. Altman, “Analysis by categorizing or dichotomizing continuous variables is inadvisable: an example from the natural history of unruptured aneurysms,” American Journal of Neuroradiology, vol. 32, no. 3, pp. 437–440, 2011.
[18]  C. Bennette and A. Vickers, “Against Quantiles: categorization of continuous variables in epidemiologic research, and its discontents,” BMC Medical Research Methodology, vol. 12, article 21, 2012.


comments powered by Disqus

Contact Us


微信:OALib Journal