Assessment of Clinical Variables Importance with the Use of Neural Networks by the Example of Thyroid Blood Test Parameters

  • Martyna MichałowskaEmail author
  • Tomasz Walczak
  • Jakub Krzysztof Grabski
  • Monika Grygorowicz
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 925)


Screening blood tests for thyrometabolic status determination are difficult in interpretation because of many factors like age, sex or measurement method that influence their proper interpretation. To solve this problem machine learning techniques, like artificial neural networks (ANN) can be applied, but their application is very common belittled due to their very little explanatory insight to the relation between input parameters (e.g. test results) and model of the disease. In contrast to previous studies concerning application of neural networks in thyroid disease diagnosis, in this study the authors decided to focus on extraction of reliable dataset (with preserved proportion of diseased and health cases) and quantification of input parameters importance in neural network decisive process. The importance of the variables considered as the most significant in hypothyroidism detection was estimated based on two independent methods: connection weights method (according to the Garson’s algorithm) and sensitivity analysis. The results show, that the most important factors in hypothyroidism detection are TSH, TT4, FTI and age, and the rejection of other analyzed in this study parameters (sex, T3, T4U) does not influence significantly the performance of the neural network model and its predictive power.


Hypothyroidism Neural network analysis Parameter importance 



This study was conducted with the use of database from UCI Machine Learning Repository donated by Garavan Institute in Sydney, Australia [11].

Presented research results were funded with the grant 02/21/DSMK/3529 and 02/21/DSPB/3513 allocated by the Ministry of Science and Higher Education in Poland.


  1. 1.
    Franklyn, J., Shephard, M.: Evaluation of thyroid function in health and disease. In: De Groot, L., Chrousos, G., Dungan, K., et al. (eds.) Endotext, South Dartmouth (MA) (2000)Google Scholar
  2. 2.
    Pellitteri, P.K., Ing, S., Jameson, B.: Disorders of the thyroid gland. In: Flint, P., Haughey, B.H., Lund, V., et al. (eds.) Cummings Otolaryngology, 6th edn, pp. 1884.e3–1900.e3. Elsevier Inc. (2015)Google Scholar
  3. 3.
    Arbib, N., Hadar, E., Sneh-Arbib, O., et al.: First trimester thyroid stimulating hormone as an independent risk factor for adverse pregnancy outcome. J. Matern Neonatal Med. 30, 2174–2178 (2017)CrossRefGoogle Scholar
  4. 4.
    Tadeusiewicz, R., Korbicz, J., Rutkowski, L., Duch, W.: Sieci neuronowe w inzynierii biomedycznej. Tom 9. ang. Neural Networks in Biomedical Engineering, vol. 9. Akademicka Oficyna Wydawnicza EXIT, Warszawa (2013)Google Scholar
  5. 5.
    Michałowska, M., Walczak, T., Grabski, J.K., Grygorowicz, M.: Artificial neural networks in knee injury risk evaluation among professional football players. In: AIP Conference Proceedings, Lublin, p. 70002 (2018)Google Scholar
  6. 6.
    Walczak, T., Grabski, J., Grajewska, M., Michałowska, M.: Application of artificial neural networks in man’s gait recognition. In: Advances in Mechanics: Theoretical, Computational and Interdisciplinary Issues, pp. 591–594. CRC Press (2016)Google Scholar
  7. 7.
    Grabski, J.K., Walczak, T., Michałowska, M., Cieslak, M.: Gender recognition using artificial neural networks and data coming from force plates. In: Gzik, et al., M. (eds.) Innovations in Biomedical Engineering, IBE 2017, Advances in Intelligent Systems and Computing, vol. 623, pp. 53–60. Springer, Cham (2018)Google Scholar
  8. 8.
    Zhang, G., Berardi, V.L.: An investigation of neural networks in thyroid function diagnosis. Health Care Manag. Sci. 1, 29–37 (1998)CrossRefGoogle Scholar
  9. 9.
    Temurtas, F.: A comparative study on thyroid disease diagnosis using neural networks. Expert Syst. Appl. 36, 944–949 (2009)CrossRefGoogle Scholar
  10. 10.
    Ozyilmz, L., Yildirim, T.: Diagnosis of thyroid disease using artificial neural network methods. In: Proceedings of the 9th International Conference Neural Information Process, vol. 4, pp. 2033–2036 (2002)Google Scholar
  11. 11.
    Dua, D., Karra Taniskidou, E.: UCI Machine Learning Repository. University of California, School of Information and Computer Science, Irvine (2017). Accessed 30 Mar 2018
  12. 12.
    Olden, J.D., Jackson, D.A.: Illuminating the “black box”: a randomization approach for understanding variable contributions in artificial neural networks. Ecol. Model. 154, 135–150 (2002)CrossRefGoogle Scholar
  13. 13.
    de Oña, J., Garrido, C.: Extracting the contribution of independent variables in neural network models: a new approach to handle instability. Neural Comput. Appl. 25, 859–869 (2014)CrossRefGoogle Scholar
  14. 14.
    Soldin, O.P., Soldin, D., Sastoque, M.: Gestation-specific thyroxine and thyroid stimulating hormone levels in the United States and worldwide. Ther. Drug Monit. 29, 553–559 (2007)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Martyna Michałowska
    • 1
    Email author
  • Tomasz Walczak
    • 1
  • Jakub Krzysztof Grabski
    • 1
  • Monika Grygorowicz
    • 2
  1. 1.Institute of Applied Mechanics, Faculty of Mechanical Engineering and ManagementPoznan University of TechnologyPoznańPoland
  2. 2.Department of Spondyloorthopaedics and Biomechanics of the Spine, Wiktor Dega Orthopaedic and Rehabilitation Clinical HospitalPoznan University of Medical SciencesPoznanPoland

Personalised recommendations