• S. V. Yegorov SI «Dnipropetrovsk Medical Academy of the Ministry of Health of Ukraine»
  • L. S. Koriashkina Dnipro University of Technology
  • I. Yu. Symonets Dnipro University of Technology



hypothesis verification, descriptive statistics, Ringer solution, blood test, application software


Background. Modern information technologies provide the opportunity to apply different methods of Data Mining and statistics for analyze the huge archives of medical data, containing information about the various cases of each existing disease etc.

Purpose. The purpose of this research is developing a method for a comprehensive statistical analysis of the data of medical observations for children with acute surgical pathology to detect the influence of different infusion media used in the treatment on the clinical and laboratory parameters of the children. We compare three solutions — based on glucose with the addition of electrolytes, Ringer's lactate and Ringer's malat.

Materials and methods. Input information — results of clinical observation of 137 children with acute surgical pathology aged 6 to 17 years during consecutive five days. All patients are divided into two groups by age: 1 — from 6 to 12 years, 2 — from 13 to 17 years, and into three categories by type of infusion solution used during treatment. The biochemical composition of urine and blood of each patient is represented by the following data: blood red blood cells, hemoglobin, leukocytes, diuresis speed, body surface area, daily diuresis, density and acetone in urine, potassium, sodium, chlorine, lactate, urea in blood, renal and cardiac enzymes , urea nitrogen, the rate of filtration of creatine in the kidneys; creatine, glucose and acidity of blood, total number of acidic and alkaline buffers in the blood, blood bicarbonate level, anionic gap, blood osmolarity, potassium, sodium content, creatine chloride in urine, sodium excretion by the kidneys. Instrumental researches have established: shock volume of heart, minute volume of blood, cardiac index, volume of external and intracellular fluid and circulating blood (% of body weight), total peripheral resistance of vessels.

The procedure of each mentioned indicator analysis, proposed and implemented in Java language, include: a) calculating of descriptive statistics and confidence interval for each of the five observations in each patient group; b) assessing the significance of the difference in the sample mean of two observations (the first and each subsequent one) using the student's t-criterion or Wilcox's W-criterion; c) calculating the rate of indicator change and its descriptive statistics and its comparative analysis for different solutions using the Student's t-criterion (for independent samples) or the Mann-Whitney U-criterion; d) comparison of the one therapy action in the different age patients groups.

Results. It is elicited rates for which certain therapies used in treatment have a significant or almost no effect, supporting the indicator within acceptable limits. A comparative analysis of the therapy effect in different age groups and a comparison of three types of therapy among themselves are carried out. For each of the biochemical or physiological indicators, we determined moments when a certain solution starts to affect it substantially.

Conclusions. The developed application software allows you to quickly automatically perform an analysis of the effects of various types of therapy used to treat one disease, based on patient data collected over several consecutive days. Such an analysis facilitates the generation of substantiated conclusions about further therapeutic treatment.


Samojlenko, N. E., Kuvina, V. N., Kuvin, S. S. (2009). Kompleksnyj analiz medicinskih dannyh [Comprehensive analysis of medical data]. Vestnik VGTU, 9, 1-5. [In Russian].

Hoiko, O. V. (2009). Suchasni tekhnolohii obrobky y analizu medychnykh danykh [Modern technologies processing and analysis of medical data]. Medychna informatyka ta inzheneriia (Medical informatics and engineering), 4, 39-44. [In Ukrainian].

Kachmar, V. O. (2010). Medychni informatsiini systemy — stan rozvytku v Ukraini [Medical information system — current state in Ukraine]. Ukrainskyi zhurnal telemedytsyny ta medychnoi telematyky, 8 (1), 12-17. [In Ukrainian].

Muntian, O. A., Muntian, M. L., Yarovenko, A. H. (2017). Pakety prykladnykh prohram statystychnoho analizu rezultativ naukovykh doslidzhen [Software packages for statistical analysis of research results]. Such. inf. tekhnolohii ta innovatsiini metodyky navch. u pidhotovtsi fakhivtsiv: metodolohiia, teoriia, dosvid, problemy. Zb. nauk. pr., 49, 135-138. [In Ukrainian].

Snisar', V. I., Egorov, S. V. (2016). Kristalloidnye rastvory u detej [Crystalloid solutions in children]. Bil, zneboliuvannia i intensyvna terapiia, 2, 64-76. [In Russian].

Yekhalov, V. V., Yegorov, S. V., Bagunfna, O. O. (2018). Hiponatriiemiia v ditei z khirurhichnoiu patolohiieiu

[Hyponatremia in children with surgical pathology]. Klinychna medytsyna, XXIII, 4 (1), 58-61. [In Ukrainian].

Rebrova, O. Yu. (2000). Statisticheskij analiz medicinskih dannyh. Primenenie paketa prikladnyh programm STATISTICA [Statistical analysis of medical data. Application package STATISTICA]. MediaSfera. [In Russian].

Medik, V. A., Tokmachev, M. S., Fishman, B. B. (2000). Statistika v mediczine i biologii: Rukovodstvo. V 2-kh tomakh. [Statistics in Medicine and Biology: A Guide. In 2 volumes]. Mediczina, 1, 412. [In Russian].



How to Cite

Yegorov, S. V., Koriashkina, L. S., & Symonets, I. Y. (2019). COMPREHENSIVE STATISTICAL ANALYSIS OF MEDICAL OBSERVATION DATA FOR CHILDREN WITH ACUTE SURGICAL PATHOLOGY. Medical Informatics and Engineering, (3), 69–78.