Informatics and Applications
2015, Volume 9, Issue 2, pp 63-74
MODELS FOR REPRESENTATION AND TREATMENT OF REFERENCE VALUES
Abstract
The article considers the problem of modeling reference values - results of a certain type of quantities obtained from a single individual or a group of individuals corresponding to a stated description. For this purpose, the article proposes to use a mixture of normal distributions, which can effectively serve as a means of approximating the actual data and to be accessible from the standpoint of theoretical analysis. In estimating the parameters of mixture of distributions, the major role is played by the maximum likelihood method and its embodiment in the form of the expectation-maximization (EM) algorithm. For assessing the number of mixture components, the article suggests to use the likelihood ratio test and a method based on the chi-square distance between the distributions. Their properties are investigated using the bootstrap method. As an experiment, the article considers the description of the empirical distribution of patient data, including the age and measurements of PSA (Prostate-Specific Antigen). The proposed solutions have clear advantages: high detail by age, smoothing the results of observations for age groups which are different in size, and the opportunity to form assumptions about the nature of the relationship between age and PSA.
[+] References (14)
- Sokal, R. R., and P. H. A. Sneath. 1963. Principles of nu-merical taxonomy. San Francisco - London: W. H. Free-man and Co. 359 p.
- Burtis, C. A., and D. E. Bruns. 2014. Tietz fundamentals of clinical chemistry and molecular diagnostics. St. Louis, MO: Elsevier Health Sciences. 1104 p.
- Sivkov, A., N. Keshishev, M. Krivenko, G. Kovchenko, and L. Nikonova. 2014. Comparison of chromogranin-A levels determined by different test systems in patients with prostate diseases. Eur. Urol. Suppl. 13(5):143.
- Krivenko, M. P. 2014. Sravnitel'nyy analiz protsedur re- gressionnogo analiza [Comparative analysis of regression analysis procedures]. Informatika i ee Primeneniya - In-form. Appl. 8(3):70-78.
- Harris, E. K., and J. C. Boyd. 1995. Statistical bases of reference values in laboratory medicine. New York, NY Marcel Dekker. 361 p.
- McLachlan, G., and D. Peel. 2000. Finite mixture models. New York, NY: Wiley & Sons. 456 p.
- Korolev, V. Yu. 2007. EM-algoritm, ego modifikatsii i ikh primenenie k zadache razdeleniya smesey veroyatnostnykh raspredeleniy. Teoreticheskiy obzor [EM-algorithm. Modi-fications and their application to the separation of mixtures
of probability distributions. Theoretical review]. Moscow: IPI RAN. 94 p.
- McLachlan, G., andT. Krishnan. 2008. The EM algorithm and extensions. Hoboken, NJ: Wiley & Sons. 359 p.
- Krivenko, M.P. 2011. Prikladnye metody otsenivaniya raspredeleniya mnogomernykh dannykh maloyvyborki [Ap-plied methods for estimating the distribution of multi-dimensional data of small sample size]. Moscow: IPI RAN. 146 p.
- Vasil'ev, V. G., and M. P. Krivenko. 2008. Metody avtoma- tizirovannoy obrabotki tekstov [Methods for automatic text processing]. Ě.: IPI RAN. 305 p.
- Seidel, W., K. Mosler, and M. Alker. 2000. A cautionary note on likelihood ratio tests in mixture models. Ann. Inst. Statist. Math. 52(3):481-487.
- Lehmann, E. L., and J. P. Romano. 2005. Testing statistical hypotheses. New York, NY: Springer. 792 p.
- PSA Measurements. British Association of Urological Surgeons (BAUS). March 2014. Available at: http:// www.baus.org.uk (accessed February 11, 2015).
- Brosman, S. A. January 13, 2015 (updated). Age- specific PSA reference ranges. 2015. Available at: http:// emedicine.medscape.com/article/457394-overview (ac-cessed February 11, 2015).
[+] About this article
Title
MODELS FOR REPRESENTATION AND TREATMENT OF REFERENCE VALUES
Journal
Informatics and Applications
2015, Volume 9, Issue 2, pp 63-74
Cover Date
2015-02-30
DOI
10.14357/19922264150208
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
mixture of normal distributions; assessing the number of components in mixture models; reference values
Authors
M. P. Krivenko
Author Affiliations
Institute of Informatics Problems, Federal Research Center “Computer Science and Control” of the Russian
Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|