Creation of Mathematical Model to Optimize Breast Cancer Screening Program

A.V. Rusyn, V.I. Rusyn, O.M. Odoshevska, O.T. Devinyak

Abstract


This paper describes the work on optimization of breast cancer screening questionnaire, mathematical model created on its base, which is capable of determining the risk of breast cancer developing; there is identified the impact of factors of history and lifestyle on cancer risk.
Materials and Methods. Statistical analysis and modeling were performed in R 3.0.1. L1-regularized logistic regression model was used to determine the formula for calculating the risk of breast cancer, the operating characteristics analysis of the model was carried out in order to optimize the border between classes.
Results and Discussion. Based on a survey of 321 women, questionnaire for breast cancer screening has been optimized, the negative impact of main factors on the risk of breast cancer is confirmed. In addition to high-risk factors a number of favorable factors that reduce the risk of cancer have been identified: breast-feeding for more than 3 months, ischomenia before 45 years. The new model for determining breast cancer risk based on a survey of the female population is characterized by high accuracy of fitting (97.5 %) and prediction (94.1 %), that made it possible to create a computer program for use in clinics, that is capable to determine the risk of breast cancer.
Conclusion. Using proposed mathematical modeling significantly improves the efficiency of questionnaire screening, makes survey easy due to reducing the number of questions and appears to be more accurate and fast way to determine groups at risk of breast cancer.


Keywords


risk factors; breast cancer; prediction model; mathematical modeling

References


Білинський Б.Т. Еволюція клінічних підходів до проблеми раку грудної залози на фоні прогресу онкологічної науки / Б.Т. Білинський // Онкологія. — 2010. — Т. 12, № 3. — С. 282-285.

Искусственные нейронные сети: прогнозирование вероятности развития рака молочной железы у женщин, имеющих факторы риска / [Ю.В. Думанский, В.В. Приходченко, Ю.Е. Лях, В.Г. Гурьянов] // Нейронауки: теоретичні та клінічні аспекти. — 2007. — Т. 3, № 1–2. — С. 106-109.

Приходченко В.В. Анкетный скрининг как метод первичного отбора групп риска заболеваний молочной железы (предварительное сообщение) / В.В. Приходченко // Медико-соціальні проблеми сім’ї. — 2007. — Т. 12, № 1–2. — С. 57-65.

Профилактика рака молочной железы / В.Ф. Семиглазов, Г.А. Дашян, В.В. Семиглазов // Практическая онкология. — 2011. — Т. 12, № 2. — С. 66-69.

Смоланка І.І. Профілактика і рання діагностика раку молочної залози / І.І. Смоланка, С.Ю. Скляр, І.І. Досенко // Жіночий лікар. — 2009. — № 5. — С. 40-45.

Факторы риска злокачественных и доброкачественных заболеваний молочной железы / И.А. Коноплева, В.Ф. Левшин, Е.Г. Пинносевич [и др.] // Советская медицина. — 1990. — № 12. — С. 93-96.

Харченко В.П. Скрининг и возможности совершенствования ранней диагностики рака молочной железы / В.П. Харченко, Н.И. Рожкова, Е.В. Меских // Вестник Московского онкологического общества. — 2006. — № 11. — С. 4-5.

Assessment of the accuracy of the Gail model in women with atypical hyperplasia / V.S. Pankratz, L.C. Hartmann, A.C. Degnim [et al.] // J. Clin. Oncol. — 2008. — Vol. 26(33). — P. 5374-5379.

Breast cancer risk assessment in the Czech female population — an adjustment of the original Gail model / J. Novotny, L. Pecen, L. Petruzelka [et al.] // Breast Cancer Res Treat. — 2006. — Vol. 95. — P. 29-35.

Claus E.B. Autosomal dominant inheritance of early onset breast cancer / E.B. Claus, N. Risch, W.D. Thompson // Cancer. — 1994. — Vol. 73. — P. 643-651

Friedman J. Regularization Paths for Generalized Linear Models via Coordinate Descent / J. Friedman, T. Hastie, R. Tibshirani // Journal of Statistical Software. — 2010. — 33(1). — P. 1-22.

Pan W. Akaike’s information criterion in generalized estimating equations / W. Pan // Biometrics. — 2001. — 57(1). — P. 120-125.

Projecting absolute invasive breast cancer risk in white women with a model that includes mammographic density / J. Chen, D. Pee, R. Ayyagari [et al.] // J. Natl. Cancer. Inst. — 2006. — Vol. 98. — P. 1215-1226.

Projecting individualized probabilities of developing breast cancer for white females who are being examined annually / M.N. Gail, L.A. Brinton, D.P. Byar [et al.] // J. Natl. Cancer Inst. — 1989. — Vol. 81. —

P. 1879-1989.

Proportion of breast cancer cases in the United States explained by well-established risk factors / M.P. Madigan, R.G. Ziegler, J. Benichou [et al.] // J. Natl. Cancer. Inst. — 1995. — Vol. 87(22). — P. 1681-1685.

Pu X. Development and validation of risk models and molecular diagnostics to permit personalized management of cancer / Xia Pu, Y. Ye, X. Wu // Cancer. — 2014. — Vol. 120, Issue 1. — P. 11-19.

Risk prediction models of breast cancer: a systematic review of model performances / T. Anothaisintawee,

Y. Teerawattananon, N. Wiratkapun [et al.] // Breast Cancer Res Treat. — 2012. — Vol. 133. — P. 1-10.

ROCR: visualizing classifier performance in R / T. Sing, O. Sander, N. Beerenwinkel, T. Lengauer // Bioinformatics. — 2005. — Vol. 21(20). — P. 3940-3941.

The BOADICEA model of genetic susceptibility to breast and ovarian cancers: updates and extensions / A.C. Antoniou, A.P. Cunningham, J. Peto [et al.] // Br. J. Cancer. — 2008. — Vol. 98(8). —

P. 1457-1466.

Tyrer J. A breast cancer prediction model incor­porating familial and personal risk factors / J. Tyrer, S.W. Duffy, J. Cuzick // Stat Med. — 2004. — Vol. 23(7). — P. 1111-1130.

Validation studies for models pro­jecting the risk of invasive and total breast cancer incidence / J.P. Costantino, M.H. Gail, D. Pee [et al.] // J. Natl. Cancer. Inst. — 1999. — Vol. 91(18). — P. 1541-1548.

Risk prediction models for colorectal cancer: a review / A.K. Win, R.J. Macinnis, J.L. Hopper, M.A. Jenkins // Cancer. Epidemiol. Biomarkers Prev. — 2012. — Vol. 21. — P. 398-410.




DOI: https://doi.org/10.22141/1997-2938.2.25.2014.83067

Refbacks

  • There are currently no refbacks.


Copyright (c) 2016 UKRAINIAN JOURNAL OF SURGERY

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

 

© Publishing House Zaslavsky, 1997-2017

 

 Яндекс.МетрикаSeo анализ сайта Рейтинг@Mail.ru